APOL1-Associated Kidney Disease

Apolipoprotein L1 G1 and G2 high-risk genotypes as a cause of focal segmental glomerulosclerosis, collapsing glomerulopathy, HIV- and COVID-19-associated nephropathy, chronic kidney disease attributed to hypertension and early kidney allograft failure, with incomplete penetrance dependent on second hits.

Literature and regulatory information updated: 6 November 2026.

0. Initial Reference Table

1. Introduction

APOL1-associated kidney disease is the entity that resulted from the identification, in 2010, of two common coding variants of the APOL1 gene as the cause of much of the excess chronic kidney disease observed in people with recent African ancestry. It differs from classical genetic kidney diseases in two decisive respects: the responsible variants are common, not rare, and their effect is expressed through a toxic gain of function with low penetrance, dependent on additional triggering factors, so the genetic test result is not a diagnosis in itself and does not generate recurrence risks in the Mendelian sense [1,6].

1.1. The path of discovery: from MYH9 to APOL1

The path to the correct gene is instructive and explains why the literature before 2010 attributes the entity to another gene. In 2008, Kopp and colleagues performed a genome-wide admixture linkage disequilibrium mapping scan in 190 African Americans with focal segmental glomerulosclerosis and 222 controls, identifying in the 22q12 region a signal with a genome-wide lod score of 9.2 and a peak of 12.4 centred on MYH9, the gene for a non-muscle myosin heavy chain expressed in podocytes; haplotypes of exons 14 to 23 were associated with the disease with an odds ratio of 5.0 (95% confidence interval [CI] 3.5–7.1; p = 4 × 10⁻²³; n = 852), extending to end-stage kidney disease attributed to hypertension (odds ratio 2.2; 95% CI 1.5–3.4) but not to diabetic end-stage kidney disease [8]. In the same year, Kao and colleagues independently confirmed, in 1,372 cases of end-stage kidney disease and 806 controls, the association between excess African ancestry in the 22q12 region and non-diabetic end-stage kidney disease (lod score 5.70), absent in the diabetic form, with a relative risk of 0.50 (95% CI 0.39–0.63) for each copy of the European ancestral allele [8].

No causal mutations in MYH9 were ever found, however. Genovese and colleagues noted that the signature of natural selection in the region extended beyond MYH9, encompassing the APOL gene cluster, and reanalysed 205 African Americans with biopsy-proven focal segmental glomerulosclerosis and no family history, compared with 180 African American controls. They identified two independent variants in the last exon of APOL1: odds ratio of 10.5 (95% CI 6.0–18.4) for focal segmental glomerulosclerosis and 7.3 (95% CI 5.6–9.5) for end-stage kidney disease. The previous association with the MYH9 E-1 haplotype disappeared after controlling for the APOL1 variants, being explained by linkage disequilibrium: the E-1 haplotype is present on 89% of chromosomes carrying G1 and 76% of those carrying G2 [2,5]. In the same year, Tzur and colleagues arrived at the same variants from an end-stage kidney disease cohort and, since they were in almost perfect linkage disequilibrium, proposed treating them as a «missense risk haplotype» [26].

The evolutionary explanation emerged at the same time. Apolipoprotein L1 is the active component of the trypanosome lytic factor of human serum, and only the kidney disease-associated variants lyse Trypanosoma brucei rhodesiense, the subspecies responsible for the acute form of East African sleeping sickness, which is normally resistant to the ancestral form of the protein. The selective advantage against human African trypanosomiasis drove the alleles to high frequencies in sub-Saharan Africa, where they arose more than ten thousand years ago, and the African diaspora carried them to the Americas and Europe [2,7,27].

1.2. Subsequent milestones

Kopp and colleagues defined, in 2011, the magnitude of the effect by phenotype and the limits of its clinical interpretation, with recessive odds ratios of 17 (95% CI 11–26) for focal segmental glomerulosclerosis and 29 (95% CI 13–68) for HIV-associated nephropathy, and with the first estimate of lifetime risk, about 4% for focal segmental glomerulosclerosis, which established from the outset the notion of low penetrance [17]. In 2013, Parsa and colleagues showed, in the AASK (African American Study of Kidney Disease and Hypertension) and CRIC (Chronic Renal Insufficiency Cohort) studies, that the variants explain a substantial part of the faster progression of chronic kidney disease in Black patients, with and without diabetes [28]. In 2015, Larsen and colleagues described the histopathology of the subnephrotic form, until then subsumed under the label of arterionephrosclerosis [3]. In 2020, the COVID-19 epidemic revealed a new second hit, with collapsing glomerulopathy in Black patients with a high-risk genotype and no evidence of direct viral infection of the kidney, which led to the proposal of the term COVID-19-associated nephropathy [29]. In 2023, the first trial of a mechanism-targeted therapy demonstrated a reduction in proteinuria with inaxaplin [21]. In 2025, the H3Africa network study of 8,355 participants from Ghana and Nigeria quantified the risk in the ancestral region and demonstrated, robustly for the first time, an effect of a single risk allele [7]. Between 2024 and 2026, the description of the protective modifier p.(Asn264Lys), which almost completely abolishes the risk conferred by G2, was consolidated [16,19].

1.3. Nomenclature and definition

Terminology was the subject of explicit debate at the KDIGO controversies conference held in Accra, Ghana, in April 2024, with 43 professionals from various disciplines and 11 participants from West Africa and the United States representing patients, families and carers. Participants considered «APOL1-mediated», «APOL1-induced» and «APOL1-associated kidney disease», and the selection criteria were usefulness for all stakeholders, ease of use for patients and communities, flexibility to accommodate future knowledge and the ability to encompass the whole spectrum. The designations «mediated» and «associated» did not obtain universal support, because high-risk genotypes are probabilistic and not deterministic risk factors and most carriers do not develop clinically apparent disease. The designation with the greatest support was «APOL1-associated kidney disease», usable on its own, to trigger the use of targeted therapies when they exist, or as a category with subsequent specification, for example «APOL1-associated kidney disease, focal segmental glomerulosclerosis». Use of the term presupposes a role for genetic testing, but does not dispense with full diagnostic investigation, including kidney biopsy when indicated [1].

The 2021 North American Delphi consensus used the designation «APOL1-associated nephropathy» for any nephropathy in a person with a high-risk genotype and no other known cause of kidney disease, accepting that the entity overlaps with other forms of kidney disease and describing the most frequent presentation as glomerulosclerosis accompanied by tubulointerstitial and vascular changes. The same panel recommended, with 93% agreement, that classifications recognise the entity as an adjunct to the clinical diagnosis, for example «focal segmental glomerulosclerosis, APOL1 variant», a format that anticipated the category with specification adopted in 2024 [30].

The application of Hill's causality criteria, carried out at the same conference, clarifies the nosological status of the entity. The strength of association is large for common variants and reproducible across a spectrum of phenotypes with continuous and categorical outcomes, which cuts across the traditional histopathological classification and includes immune causes, such as HIV-associated nephropathy and COVID-19-associated nephropathy, and non-immune causes, such as chronic kidney disease attributed to hypertension. Consistency is established by replication in multiple cohorts, with the caveat that most carriers do not become ill. Specificity is low, because HIV-associated nephropathy, COVID-19-associated nephropathy and focal segmental glomerulosclerosis occur without the high-risk genotype. Temporality is documented in kidney disease induced by therapeutic interferon, in COVID-19-associated nephropathy and in murine models. The biological gradient is demonstrated by dose-dependent cytotoxicity in human embryonic kidney cells and by dose-dependent disease in mice with podocyte expression of the risk variants. Plausibility and coherence derive from cellular and animal data, and the experimental criterion is supported by transgenic mice that develop albuminuria and azotaemia after interferon-gamma and by the phase 2 trial that reduced proteinuria with targeted treatment. No analogy was identified [1].

1.4. Operational definition and classification consequences

For practical purposes, APOL1-associated kidney disease is defined as kidney disease, clinically or histologically characterised, in a person with a high-risk APOL1 genotype and no sufficient alternative cause, with explicit recognition that the genotype does not prove individual causality and that the presence of the protective modifier p.(Asn264Lys) in cis with G2 requires another explanation to be sought [1,16,20]. The high-risk genotype is the set G1/G1, G2/G2 or G1/G2; the presence of a single risk allele, previously considered benign, confers a measurable, albeit smaller, risk, at least in West African populations [7].

Administrative classification followed this consolidation. Previously, the absence of a dedicated code forced the misclassification of many patients, especially people of African ancestry without diabetes, and the lack of a diagnosis of APOL1-mediated kidney disease in the United States Renal Data System and in international registries confused the epidemiology of kidney failure and delayed research. The US ICD-10-CM now includes code N07.B, «hereditary nephropathy, not elsewhere classified, with APOL1-mediated kidney disease», with the inclusion terms «AMKD with glomerulonephritis» and «AMKD with glomerulosclerosis», and code Z84.11 for a family history of the same entity, which may support a request for genetic testing [4,11].

2. Molecular Genetics and Pathophysiology

2.1. The gene and the protein

APOL1 is located at 22q12.3, between coordinates chr22:36,253,133 and 36,267,525 of the GRCh38 assembly, has seven exons spread over about 14 kb and belongs to a cluster with APOL2, APOL3 and APOL4 resulting from tandem duplication, with APOL1 at the telomeric end and in the opposite orientation to the others; the promoters of the cluster lack a TATA box and use multiple transcription start sites. The clinical reference transcript is NM_003661.4 [5,6,31].

The protein is a human-specific serum apolipoprotein, carried in a subfraction of high-density lipoproteins in complex with apolipoprotein A-I. The complementary DNA cloned by Duchateau and colleagues in 1997 encodes a 383-amino-acid polypeptide with a 12-residue secretory signal peptide; the genomic analysis of Page and colleagues predicts a 398-amino-acid protein with a calculated mass of 43.9 kD. There is a minor splicing variant that includes exon 2 and predicts a 43-residue signal peptide, whereas the most common transcript, lacking that exon, predicts a 27-residue signal peptide — a relevant detail because residue numbering differs between descriptions and the historical nomenclature of the risk variants (S342G, I384M, deletion of N388 and Y389) is the most widely used in the clinical literature, without automatic correspondence with other conventions [5].

Expression is broad. Transcript analysis revealed expression in a wide variety of tissues, with the exception of fetal brain, highest in lung, spleen, prostate and placenta, and, by Northern blot, highest in placenta, lung and liver. Hepatocytes constitutively express and secrete the protein into the circulation. In human vascular tissue there is expression in endothelial cells and possibly in macrophages, and in human umbilical vein endothelial cells stimulation with tumour necrosis factor alpha increased expression tenfold. In kidneys, single-cell expression data, chromatin accessibility data and immunohistochemistry in patients with COVID-19-associated nephropathy showed APOL1 messenger RNA in peritubular and glomerular endothelial cells and in podocytes. Expression and function in distal tubular cells, immune cells and vascular smooth muscle await confirmation, and, although the RNA is detectable in several extrarenal tissues, it is not known whether splicing isoforms or measurable protein exist in those tissues [1,5].

2.2. Physiological function: innate immunity against trypanosomes

The best-established function of apolipoprotein L1 belongs to innate immunity. Vanhamme and colleagues demonstrated in 2003 that it is the trypanosome lytic factor of normal human serum: depletion of the protein, by incubation with the serum resistance-associated protein (SRA) of T. b. rhodesiense or with specific antibodies, completely abolishes trypanolytic activity, which is restored by adding native or recombinant protein. The protein is taken up by the endocytic pathway to the parasite's lysosome, and the N-terminal alpha helix of SRA interacts strongly with a carboxy-terminal alpha helix of apolipoprotein L1, conferring resistance [5].

Perez-Morga and colleagues established the molecular mechanism in 2005: the protein contains a pore-forming domain functionally similar to that of bacterial colicins, flanked by a membrane-addressing domain; in lipid bilayers it forms anion channels; in T. brucei it targets the lysosomal membrane, depolarises it, causes a continuous influx of chloride and osmotic swelling of the lysosome until the parasite lyses [5]. Together with haptoglobin-related protein (HPR), apolipoprotein L1 forms trypanosome lytic factor 1; the HPR gene is copy-number variable, with duplications at polymorphic frequencies of up to 15% in West and Central Africa, overlapping the endemic area of chronic human trypanosomiasis, and these duplications were under-transmitted to children affected by trypanosomiasis when assessed together with APOL1 alleles [5].

2.3. The G1 and G2 risk variants

The two risk variants lie in the last exon, in the SRA-interacting domain, and are mutually exclusive owing to physical proximity, with recombination between them very unlikely [2,5].

Functional characterisation explains why these variants became fixed. Both plasma from carriers and recombinant protein with G1 or G2 lyse T. b. rhodesiense, with and without SRA, although they do not lyse T. b. gambiense; the I384M substitution and the deletion of N388 and Y389 lie precisely at the SRA-binding site in the carboxy-terminal helix, so the variants escape the parasite's resistance mechanism [2,5].

The geographical distribution of the alleles is a consequence of this history. In the original analysis, the frequency of the G1 allele was 52% in 205 cases of focal segmental glomerulosclerosis and 18% in 180 African American controls (p = 1.07 × 10⁻²³), and G1 was present on about 40% of Yoruba chromosomes and on no European, Japanese or Chinese reference chromosome; G2 was detected in three Yoruba participants and in no European, Japanese or Chinese participant [2,5]. Sequencing of 1.4 kb of the last exon in 187 individuals from ten African populations identified 38 variants, including 15 non-synonymous substitutions and the 6-base-pair deletion, distributed across the pore-forming (3), membrane-addressing (5) and SRA-interacting (6, including G1 and G2) domains; G2 frequencies were similar in all populations (3–8%), whereas G1 was common only in the Yoruba of West Africa (39%). The same study described a third haplotype, designated G3, present in all populations except the Yoruba, under strong selective pressure in the Fulani, pastoralists historically exposed to T. b. gambiense, and weaker in East African populations exposed to T. b. rhodesiense [5].

2.4. Mode of transmission: recessive, with a minor monoallelic effect

The model demonstrated in the original data is recessive: comparing participants with zero or one risk allele with those who had two, the odds ratio for focal segmental glomerulosclerosis was 10.5 (95% CI 6.0–18.4), compatible with a completely recessive pattern [2,5]. The high-risk genotype therefore corresponds to G1/G1, G2/G2 or G1/G2, and for more than a decade the carrier of a single allele was considered essentially unaffected [6].

This position has been revised. The H3Africa network study, of 8,355 participants from Ghana and Nigeria, demonstrated a dose-response relationship: compared with G0/G0, carriers of one risk allele had an adjusted odds ratio for chronic kidney disease of 1.18 (95% CI 1.04–1.33) and for focal segmental glomerulosclerosis of 1.61 (95% CI 1.04–2.48), while carriers of two alleles had 1.25 (95% CI 1.11–1.40) and 1.84 (95% CI 1.30–2.61). This result changes the risk classification of carriers of a single copy, especially in the presence of genetic or environmental modifiers [7]. The KDIGO consensus recognises the emerging evidence of monoallelic risk and also mentions cohort data suggesting that, even without proteinuria, there is a higher risk of kidney failure in high-risk carriers [1]. It should be noted that the OMIM gene-phenotype table annotates entry 612551 as autosomal dominant; this is a formal database annotation, which does not correspond to the recessive model demonstrated in the primary data or to the additive monoallelic effect described later [2,7,8].

2.5. Mechanism of injury: toxic gain of function

The mechanism is not a deficiency. The variant protein exerts a toxic gain of function, which is why the language of loss of function does not apply: there is no biochemical deficit to measure, there is no serum, urinary or functional assay of variant status, and the heterozygote is essentially spared in terms of monogenic disease [6]. Apolipoprotein L1 acts as a channel or pore in trypanosome cells and in eukaryotic cells, and the G1 and G2 variants alter this function, increasing pore activity; the same property that enhances defence against the parasite becomes harmful to kidney cells [1].

Direct pharmacological evidence comes from the development of inaxaplin: the molecule selectively inhibited APOL1 channel function in HEK293 human embryonic kidney cells with tetracycline-inducible expression and reduced proteinuria in a transgenic mouse model homologous to G2, which links channel activity to the mechanism of injury [6,21].

In cellular systems, G0 is generally non-toxic and G1 and G2 behave as toxic gain-of-function variants, with partly distinct effects in podocytes, HEK293 cells and pulmonary endothelial cells; cell death is the usual readout, although it is not established whether it is the relevant event in patients. In mice, transgenic expression of human APOL1 reproduces key features of human pathology: expression of G0 is not associated with phenotypic changes, except in one model with a pre-eclampsia-like syndrome, whereas expression of G1 or G2 in podocytes induces proteinuria and glomerular lesions resembling focal segmental glomerulosclerosis, and transgenic mice with the endogenous human promoter develop albuminuria and azotaemia after interferon-gamma. An important limitation of the models is that they reproduce rapidly progressive proteinuric glomerular disease, which corresponds to a minority of human disease, and there is no model of the slowly progressive, low-proteinuria form, which is the vast majority [1].

The downstream pathways implicated include ion channel activity, changes in the actin cytoskeleton, mitochondrial dysfunction, endoplasmic reticulum stress, reduced global protein synthesis, integrin activation dependent on the soluble urokinase-type plasminogen activator receptor, activation of nucleotide sensors and inflammation. Although podocyte expression of the variants is necessary for glomerulosclerosis, data are accumulating on extrapodocyte roles, namely endothelial and immune [1].

2.6. Incomplete penetrance and second hits

The most important clinical feature from a genetic point of view is low penetrance. It is estimated that about 15% of carriers of a high-risk genotype will develop kidney failure and that 5% to 8% will develop focal segmental glomerulosclerosis; the lifetime risk of focal segmental glomerulosclerosis estimated from the 2011 data is about 4%. The dominant hypothesis is that carriers need additional injurious factors, termed second hits [1,16,17].

The size of the second hit determines the outcome. In untreated HIV infection, the risk of HIV-associated nephropathy in carriers of a high-risk genotype rises to about 50%, and in African Americans two risk alleles explain 18% of focal segmental glomerulosclerosis and 35% of HIV-associated nephropathy [6,17]. The non-genetic second hits identified, validated and replicated include high-interferon states — with the strongest evidence —, viral infections (HIV and SARS-CoV-2) and therapeutic exposure to interferon alpha, beta or gamma, all converging on increased APOL1 expression; expression is also regulated by inflammation, hypoxia and ischaemia and by nucleic acids [1].

The genetic modifiers described include UBD, AHDC1, GSTM1 and SMOC2 as possible contributors to progression, and a truncating APOL3 variant, rs1108978, which was associated with a higher risk of chronic kidney disease in monoallelic carriers of G1 or G2, suggesting epistatic interaction and a protective effect of wild-type APOL3 [1]. Identification of modifiers is hampered by the broad clinical spectrum, by the possibility that different toxicity mechanisms have different modifiers, by confounding with common initiating events, such as diabetes or lupus, and by small sample sizes in the forms in which the association with incident disease is clearest [1].

2.7. The protective modifier p.(Asn264Lys)

The most robust genetic modifier is the APOL1 variant c.792C>A, p.(Asn264Lys), designated M1 or N264K, classified in ClinVar as benign or likely benign and with the identifier rs73885316 [19,31]. It exists in two configurations: on a fraction of the G0 haplotype, non-risk and more common, and on a fraction of the G2 haplotype; it does not occur on the G1 haplotype [19].

When co-inherited in cis with a G2 allele, it practically abolishes the risk effect, reducing it to levels comparable to G0. In high-risk cohorts with focal segmental glomerulosclerosis compared with high-risk controls without kidney disease, the reduction in the odds ratio was 93% (odds ratio 0.07), and in G2/G2 genotypes with the variant no case of focal segmental glomerulosclerosis was observed (odds ratio 0; 95% CI 0–0.14) [19]. In the discovery cohort of Martinelli and colleagues, which combined 107,696 individuals from two hospital biobanks, among carriers of a high-risk genotype the variant was inversely associated with chronic kidney disease (15 of 762 [1.97%] vs 26 of 651 [3.99%] in controls; Fisher's exact odds ratio 0.48; 95% CI 0.24–0.96; p = 2.62 × 10⁻²), an effect driven by the subgroup with focal segmental glomerulosclerosis or steroid-resistant nephrotic syndrome (2 of 320 cases, 0.63%). In low-risk genotypes there was no independent protective effect, including in an analysis restricted to G0/G0, which indicates that the protection is specific to the G2 context. Validation in eMERGE-III, UK Biobank and All of Us showed a concordant but non-significant effect, owing to the scarcity of focal segmental glomerulosclerosis cases in these population biobanks [16].

The mechanism has not been elucidated. Reduced plasma membrane permeability with inhibition of toxicity and interference with actomyosin control by APOL3 in the podocyte have been proposed, rather than specific inhibition of channel activity; structural studies suggest occlusion of a pore at the cell surface [19]. Experimental co-transfection data in HEK-293 cells modelling heterozygosity showed that the variant reduces G2-dependent cytotoxicity only in cis, in the same construct as G2, and not in trans, on the G0 allele, with comparable protein expression; the effect was stronger on one haplotype background than on another. This observation has a diagnostic consequence: when a report indicates both G2 and p.(Asn264Lys) without phase information, the risk interpretation is indeterminate [32].

The frequency of the variant limits its role as a general explanation for incomplete penetrance: it is present in less than 1% of people of African ancestry [1]. In the genotypes where it matters, estimates vary: in the Million Veteran Program, with 121,492 participants of African ancestry, 579 carriers were observed among 9,695 carriers of G1/G2 or G2/G2, about 6%, and 0.4% in the total cohort; in a French routine-practice series, a single carrier was found among 31 patients with G1/G2 and G2/G2 genotypes, corresponding to an allele frequency of 3.2% [33].

3. Clinical Spectrum and Manifestations by System

3.1. Overview of the spectrum

APOL1-associated kidney disease has no single presentation. The spectrum encompasses continuous disease phenotypes, such as albuminuria and decline in estimated glomerular filtration rate, and categorical phenotypes defined by histology or context, such as focal segmental glomerulosclerosis, collapsing glomerulopathy, HIV-associated nephropathy, COVID-19-associated nephropathy and chronic kidney disease attributed to hypertension. The typical shared features are recent African ancestry, within the last ten thousand years, variable proteinuria, interstitial fibrosis and reduced kidney function. Even among carriers of a high-risk genotype it is not possible to predict who will develop chronic kidney disease, idiopathic focal segmental glomerulosclerosis, disease attributed to hypertension, HIV-associated nephropathy, or early failure of a graft from a donor with a high-risk genotype [1].

The manifestations described are renal and obstetric. There is no established extrarenal disease: an individual-participant meta-analysis of 21,305 Black people from eight cohorts, with 8.9 ± 5.0 years of follow-up and 2,076 incident cardiovascular events among the 16,216 participants without cardiovascular disease at entry, found no association between two risk alleles and incident cardiovascular disease (hazard ratio 1.11; 95% CI 0.96–1.28), nor with coronary disease, myocardial infarction, stroke or heart failure separately, nor with mortality, and with no differences by kidney function, age, diabetes or body mass index. The interpretation is that the variants do not appear to have a direct cardiovascular effect independent of the renal effect [6,34].

3.2. Focal segmental glomerulosclerosis and collapsing glomerulopathy

Focal segmental glomerulosclerosis is the phenotype with the strongest association. In a series of 271 African American cases, 168 European American cases and 939 controls, the recessive model gave an odds ratio of 17 (95% CI 11–26), and among African Americans 40% to 60% of patients with primary focal segmental glomerulosclerosis have two risk alleles [1,17]. The form associated with two alleles is distinguished by its natural history: earlier age at onset (p = 0.01) and faster progression to end-stage kidney disease (p < 0.01), with steroid sensitivity similar to that of other patients — an important distinction, because faster progression does not translate into greater steroid resistance [17].

Collapsing glomerulopathy is the most severe form of the spectrum, characterised by segmental or global collapse of the glomerular capillary tuft with marked hyperplasia of the overlying epithelial cells, extensive foot process effacement and an often rapid course towards the need for kidney replacement therapy. In the Brazilian series of collapsing glomerulopathy previously classified as idiopathic, causal genetic variants were detected in 58.6% of patients, of which 80.5% corresponded to high-risk APOL1 genotypes and 19.5% to Mendelian variants, which redefines most of this histological category as disease of known cause [14,35].

3.3. Chronic kidney disease attributed to hypertension and the subnephrotic form

A substantial fraction of patients have neither nephrotic syndrome nor focal segmental glomerulosclerosis, but slowly progressive chronic disease with modest proteinuria, historically labelled hypertensive nephrosclerosis. In the AASK trial, of 693 Black patients with chronic kidney disease attributed to hypertension, 160 (23%) had two risk alleles, and in these the composite outcome of end-stage kidney disease or doubling of serum creatinine occurred in 58.1%, vs 36.6% in the low-risk group (hazard ratio 1.88; p < 0.001), with no interaction with the trial interventions or with the presence of baseline proteinuria [28].

The histopathology of this form was characterised in 196 African Americans with chronic kidney disease without nephrotic syndrome who underwent biopsy for suspected arterionephrosclerosis. In the discovery phase, comparing 58 patients with two alleles and 56 with no risk alleles, carriers of two alleles had less obsolescent-type global glomerulosclerosis (3.0 ± 4.3, corresponding to 31% of sclerotic glomeruli, vs 5.3 ± 6.5, equivalent to 56%; p = 0.001), more solidified and «disappearing»-type glomerulosclerosis, more thyroidisation-type tubular atrophy and more microcystic tubular dilatation, whereas patients without risk alleles had greater degrees of arteriosclerosis; segmental glomerulosclerosis did not differ between groups. Mean age was lower in carriers of two alleles (47 vs 56 years; p = 0.001). There were no differences in serum creatinine or degree of proteinuria between genotype groups, which explains why the clinical picture alone does not distinguish the forms [3].

3.4. HIV-associated nephropathy

HIV-associated nephropathy is the paradigmatic example of interaction between genotype and second hit. In African Americans, the recessive odds ratio was 29 (95% CI 13–68), and two alleles explain 35% of cases [17]. In antiretroviral-naive Black South Africans, the association was even stronger: among 116 genotyped patients (38 with HIV-associated nephropathy, 39 with chronic kidney disease and HIV positive, 39 with chronic kidney disease and HIV negative) and 108 controls, 79% of patients with HIV-associated nephropathy and 2% of population controls had two risk alleles, with a recessive odds ratio of 89 (95% CI 18–912; p < 0.001) compared with HIV-positive controls; population allele frequencies were 7.3% for G1 and 11.1% for G2, and there was no significant association with other forms of chronic kidney disease [36]. The risk of HIV-associated nephropathy in a carrier of a high-risk genotype with untreated infection approaches 50%, a value that contrasts with the lifetime risk of about 4% for focal segmental glomerulosclerosis in the absence of a known second hit, and that illustrates the dependence of the phenotype on the interferon stimulus [6,17].

3.5. COVID-19-associated nephropathy and other infection-triggered nephropathies

During the COVID-19 pandemic an analogous entity was described. In six Black patients with COVID-19 — four men and two women, mean age 55 years — who presented with acute kidney injury and new-onset nephrotic-range proteinuria, mean serum creatinine on the day of biopsy was 6.5 mg/dl and the mean urine protein/creatinine ratio 11.5 g/g. Biopsies showed collapsing glomerulopathy, extensive foot process effacement and focal or diffuse acute tubular injury, with endothelial reticular aggregates in three cases, with no viral particles or SARS-CoV-2 RNA detectable in the tissue. All six had a high-risk genotype. Five required dialysis, two died and one partially recovered without dialysis. The absence of direct viral infection of the kidney supports the two-hit model with a cytokine-mediated host response, and the authors proposed the designation COVID-19-associated nephropathy [29].

The principle generalises to other infections with an interferon response. Three Brazilian cases of dengue-triggered collapsing glomerulopathy, in patients with a high-risk genotype, were described as the first series of this association, in a setting in which the usual renal lesion of dengue is tubulointerstitial [37]. In the transplant setting, five cases of de novo collapsing focal segmental glomerulosclerosis occurred in kidneys from three deceased donors with two risk alleles, four of them preceded by viraemia, with cytomegalovirus in three and BK virus in one [38].

3.6. Lupus nephritis and other glomerulopathies

The risk variants modify the outcome of glomerulopathies with another primary cause. In transplant candidates, the high-risk genotype is associated with severe lupus nephritis, in addition to focal segmental glomerulosclerosis and HIV- and COVID-19-associated collapsing glomerulopathy [20]. However, in the West African study of 866 biopsies, in which the diagnoses were minimal change disease in 300 (34.6%), focal segmental glomerulosclerosis in 214 (24.7%), lupus nephritis in 101 (11.7%), membranous nephropathy in 88 (10.2%) and others in 161 (18.6%), the association with the high-risk genotype was found only for focal segmental glomerulosclerosis, not for minimal change disease, membranous nephropathy or lupus nephritis [7]. In Brazilian populations with lupus nephritis, risk allele frequencies were similar to those of controls, with a low prevalence of high-risk genotypes, which differs from African American cohorts and suggests regional differences or the contribution of additional modifiers [14]. The reconciliation of these results is that the high-risk genotype acts mainly as an accelerator of progression in already established immune diseases, rather than as the primary cause of those diseases.

3.7. Sickle cell nephropathy and sickle cell trait

Sickle cell trait and APOL1 risk genotypes are independent renal risk factors that frequently coexist in populations of African ancestry. In a population-based study from Bahia, which included 889 newborns from a population survey and 424 patients with biopsy-proven kidney disease, the high-risk genotype was more frequent in patients than in the general population (4.7% vs 1.2%; p < 0.001), whereas the frequency of sickle cell trait was the same in both groups (4.7%). In patients, the high-risk genotype was associated with greater histopathological chronicity, with more tubular atrophy and interstitial fibrosis, and progression to end-stage kidney disease increased in a graded manner with the number of risk alleles: 10%, 14% and 56% for zero, one and two alleles (p < 0.001). Progression was more frequent in carriers of sickle cell trait than in people with haemoglobin AA (39% vs 13%; p = 0.01), and in univariate analysis both two risk alleles (relative risk 4.8; 95% CI 2.8–8.4) and sickle cell trait (relative risk 2.7; 95% CI 1.4–5.1) were associated with the composite outcome of end-stage kidney disease or death [13]. It should be noted that clinical trials of targeted therapies systematically exclude sickle cell disease, so there is no evidence of efficacy in this population [39–41].

3.8. Manifestations in childhood

In the United States, children with a high-risk genotype have an increased risk of early-onset chronic kidney disease, with a more aggressive form of glomerular disease, possibly related to the high rates of focal segmental glomerulosclerosis and to prematurity. In sub-Saharan Africa, risk genotypes are likewise associated with early kidney injury, and in children and young people with perinatal HIV infection the high-risk genotype was associated with chronic kidney disease. In young adults, cross-sectional data indicate that a single copy of a risk allele is associated with higher blood pressure [1,42].

In Kinshasa, in the Democratic Republic of the Congo, in a cross-sectional study of 412 children from the general population and 401 HIV-infected children, 29 of 412 (7.0%) from the general population had a high-risk genotype, 84 (20.4%) were G1/G0 and 61 (14.8%) G2/G0; among the children with HIV, 23 of 401 (5.7%) had a high-risk genotype. In the general population, elevated albuminuria, defined as a urine albumin/creatinine ratio of 30 mg/g or more, occurred in 5 of 29 (17.2%) high-risk carriers and in 35 of 383 (9.0%) low-risk carriers (odds ratio 2.1; 95% CI 0.6–6.0; p = 0.13), a non-significant difference; in children with HIV, the high-risk genotype was associated with odds of albuminuria about 22 times higher, illustrating once again the role of the second hit [18].

With regard to childhood nephrotic syndrome, the distinction between steroid-sensitive and steroid-resistant forms is illuminating. In 115 African American children, 65 with steroid-sensitive nephrotic syndrome and 50 with the steroid-resistant form, high-risk APOL1 variants were not associated with the steroid-sensitive form (p = 0.5) and were associated with the steroid-resistant form (p = 1.04 × 10⁻⁷; odds ratio 4.17; 95% CI 2.23–7.64), whereas the HLA-DQA1 locus explained the risk of the steroid-sensitive form [43]. In Brazilian cohorts from the south-east, high-risk genotypes were associated with paediatric nephrotic syndrome with accelerated progression, even with a single allele, and are present in about 8% of paediatric cases [14].

Not all paediatric renal phenotypes are modified by the genotype. In a pilot study of 195 patients of African ancestry with congenital anomalies of the kidney and urinary tract, with a median age of 15.35 years, of whom 20 (10.3%) had a high-risk genotype, no differences were found in proteinuria (odds ratio 0.24; 95% CI 0.04–1.30), hypertension (0.99; 0.38–2.54) or chronic kidney disease stage G2 or higher (0.71; 0.29–1.78), and the median time to kidney failure was 11.59 years in high-risk carriers and 14.35 years in low-risk carriers (p = 0.73) [44].

3.9. Pregnancy: pre-eclampsia and fetal outcomes

The association with pre-eclampsia is peculiar because the relevant genotype is the fetal one and not the maternal one. In two North American cohorts, the fetal high-risk genotype was associated with pre-eclampsia with odds ratios of 1.84 (95% CI 1.11–2.93) and 1.92 (95% CI 1.05–3.49). Black infants from pregnancies complicated by pre-eclampsia were 1.8 times more likely to have a fetal high-risk genotype than the Black control population (20% vs 12%), whereas the frequency of the maternal high-risk genotype in mothers with pre-eclampsia did not differ from the general population (8% vs 12%; odds ratio 0.72; 95% CI 0.11–2.50). Excluding induced deliveries and elective caesarean sections, leaving 68 cases, the odds ratio rose to 2.09 (95% CI 1.10–3.97). The population attributable risk of the fetal high-risk genotype was estimated at 12.3%, which corresponds to about one in eight pregnancies with pre-eclampsia in Black women. Placental histopathology did not vary with fetal or maternal genotype, nor did the prevalence of classic pre-eclampsia risk factors. Among women with pre-eclampsia, all those whose fetus had a high-risk genotype had an sFlt-1/PlGF ratio above 38, vs 5 of 10 of those whose fetus had a low-risk genotype [45].

Subsequent literature extended the association to low birth weight, prematurity and small-for-gestational-age newborns. Murine models associate the variants with structural placental pathology and with activation of several cellular pathways, although clinical studies have not found a clear association with placental pathology, which suggests, here too, the need for a second hit. The question of whether the risk arises from the maternal genotype, the fetal genotype or the discordance between them remains open and has been listed as a research priority [1,46].

3.10. Living kidney donation and transplantation

The donor genotype has documented clinical consequences. In the APOLLO network cohort of North American living donors who donated between January 2000 and December 2008, assessed at home visits carried out between March 2020 and March 2024, 445 Black donors (295 women, 66%; mean age 38 years, standard deviation 10) and 208 White donors (141 women, 68%; 44 years, standard deviation 10) were included. Sixty-eight Black donors (15.3%) had a high-risk genotype. At a median of 18.5 years (interquartile range 16.9–20.5) after donation, 46 participants (7.0%) had an estimated glomerular filtration rate below 45 ml/min/1.73 m². Black donors with a high-risk genotype had a higher risk of this outcome than Black donors without a high-risk genotype (relative risk 2.31; 95% CI 1.16–4.61; p = 0.02), attenuated after adjustment for estimated glomerular filtration rate before donation (1.91; 95% CI 0.90–4.03; p = 0.09) [47].

On the recipient side, kidneys from deceased donors with a high-risk genotype are associated with faster graft failure, and results regarding the recipient genotype are mixed, although some data suggest reduced graft survival [1]. In a Brazilian review, kidneys from deceased donors with two risk variants had a worse prognosis than those from donors with no or one variant, regardless of donor ethnicity, and survival of grafts from donors of African ancestry without variants was similar to that of donors without that ancestry; kidneys from high-risk donors may already have prior injury, with podocyte density about 15% lower on pre-implantation biopsy. Unlike donor variants, whose injury is local and podocytic, recipient variants appear to relate to circulating apolipoprotein L1 [48]. In France, recipients of kidneys from deceased donors of African ancestry with two risk alleles had slightly shorter graft survival [15].

4. Clinical Recognition and Warning Signs

4.1. When to suspect, by context

Suspicion arises from the combination of recent African ancestry with a renal picture that is poorly explained by the attributed cause. The first context is chronic kidney disease labelled hypertensive nephropathy or nephrosclerosis in a person of sub-Saharan or West African ancestry: many of these patients do in fact have two APOL1 risk alleles, and reclassification is now recognised as necessary [4,20]. The second is focal segmental glomerulosclerosis, especially when it begins before the age of 40, progresses rapidly or appears in more than one family member without a clear Mendelian pattern: in a French cohort of patients with steroid-resistant nephrotic syndrome or focal segmental glomerulosclerosis of African or Antillean origin, the high-risk genotype was present in 43.1% and generally did not coexist with another causal variant in known monogenic genes, and it was observed that families with several affected individuals may have APOL1-associated disease rather than monogenic disease [15]. The third is collapsing glomerulopathy, in which the probability of APOL1-associated disease is the highest of the whole spectrum [35].

The fourth context is that of nephropathy in a patient with an infection that generates an interferon state. In any compatible nephropathy, HIV testing is essential, because untreated infection is the best-documented second hit and is treatable [6]. Acute kidney injury with new-onset nephrotic-range proteinuria during COVID-19, dengue or another systemic viral infection in a patient of African ancestry should raise the same suspicion [29,37]. Therapeutic exposure to interferon alpha, beta or gamma is a recognised trigger and should be sought in the history [1].

The fifth context is the assessment of a living kidney donor candidate of sub-Saharan or West African ancestry, regardless of family history [20]. The sixth is unexpected deterioration of graft function, especially with new-onset proteinuria, in a recipient of a kidney from a donor of African ancestry, or after cytomegalovirus or BK virus viraemia [1,38]. The seventh is pre-eclampsia in Black women, a context in which the relevant genotype is that of the fetus, with implications still under investigation [45,46].

4.2. Clinical warning signs

4.3. Emergency signs

The situations requiring immediate action are those of acute kidney disease, not specific to the entity but frequent in it: acute kidney injury with oliguria, hypervolaemia or severe electrolyte disturbances during a systemic viral infection; established nephrotic syndrome with profound hypoalbuminaemia, thromboembolic risk and anasarca; and rapid deterioration of kidney graft function. In the COVID-19-associated nephropathy series, five of the six patients required dialysis and two died, which gives a measure of the possible severity [29]. In the first patient of the Brazilian series of dengue-triggered collapsing glomerulopathy, creatinine was 11 mg/dl on admission, with proteinuria of 24.4 g in 24 hours, and the patient did not recover kidney function [37]. Pre-eclampsia is in itself an obstetric emergency, and the association with the fetal genotype does not change its management [45].

4.4. Newborn screening and population screening

There is no newborn screening for APOL1-associated kidney disease and, to date, there is no evidence to support population screening. The position of the KDIGO consensus is explicit: at population level there is insufficient evidence to recommend routine testing or screening, because there are no established treatments and the genotype is not in itself actionable, and because testing has the potential to reinforce racialised medicine, given the historical and persistent confusion between race, ethnicity and ancestry [1].

The individual position differs from the population position. For the individual person, genotyping may be a relevant clinical consideration, because the high-risk genotype is associated with a significantly higher risk of progression and because the result may inform risk stratification and the frequency of follow-up; in addition, knowledge of the genotype by the patient and clinician was associated with greater blood pressure reduction, more kidney disease screening and more self-reported behaviour change [1]. The pragmatic randomised GUARDD-US trial quantified these effects and delimited them, and its results are presented in section 5.

In children, the position is one of informed waiting: in children with focal segmental glomerulosclerosis and chronic kidney disease, genotyping may inform prognosis, and early targeted testing would allow early surveillance and intervention, but unnecessary testing is costly and potentially anxiety-provoking; when effective treatments exist, recommendations for screening and early testing in infants and children should be considered [1].

For a testing programme to be acceptable, the consensus lists operational requirements: a validated test in a registered laboratory with trained staff, pre- and post-test counselling, co-designed educational material adapted to literacy level, and an established process of early surveillance and intervention, including in childhood. In high-prevalence populations, involving community leaders, content experts and religious leaders in education initiatives is recommended [1].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Nature of the diagnosis

The diagnosis combines two complementary pieces of information that do not replace each other: the histology or clinical characterisation of the kidney disease, and the APOL1 genotype. The genetic test is definitive as to the presence of the risk alleles, but the presence of the alleles is neither diagnostic nor prognostic of any specific kidney disease [1]. Conversely, biopsy-proven focal segmental glomerulosclerosis does not confirm an APOL1-associated cause, because most focal segmental glomeruloscleroses are not related to APOL1 — two risk alleles explain about 18% of cases in African Americans [6,17]. Biopsy remains necessary to establish the lesion and to identify alternative or concomitant causes, and use of the diagnostic term does not dispense with full investigation [1,6].

The risk of a diagnostic shortcut is real and has been explicitly stated: the high frequency of the high-risk genotype, together with incomplete penetrance and limited predictive value, creates the danger of a «label diagnosis» that leads to incomplete investigation, for example avoiding an indicated kidney biopsy [16].

5.2. Decision criteria for testing

The KDIGO consensus established a list of conditions that should be met before testing is offered, presented as a set and not as alternatives [1].

The North American practice resource for transplant candidates adds two operational rules: APOL1 counselling and testing should be offered to all living donor candidates of sub-Saharan or West African ancestry, regardless of family history of kidney disease; and patients with steroid-resistant nephrotic syndrome or focal segmental glomerulosclerosis in whom suspicion of genetic disease is high, owing to early onset, positive family history or extrarenal features, should undergo genetic testing as part of the transplant evaluation, because the result, positive or negative, has implications for management before and after transplantation [20].

The 2021 Delphi consensus, the only clinical practice guidance dedicated to the entity prior to the KDIGO consensus, formulated broader and less conditional criteria. The consensus group had 14 voting members (a bioethicist from the steering committee and 13 experts, including two patient representatives and 11 specialists in adult and paediatric nephrology, cardiology, population health, medical genetics, biostatistics and bioethics; eight of African ancestry), voted in three anonymous rounds between July 2019 and February 2020 on a five-point Likert scale and defined consensus as at least 75% «agree» or «strongly agree»; the systematic review underpinning the statements covered MEDLINE and Embase from 1 January 2009 to 14 July 2020, with GRADE grading. Screening of the healthy population and testing of transplant donors and recipients were outside the scope [30].

The diagnostic statements were as follows. APOL1-associated nephropathy should be considered in all patients with kidney disease and African ancestry, especially when there is a family history of chronic kidney disease (100% agreement; GRADE 1B), and, regardless of ancestry, in all patients with kidney disease and a relative with a confirmed high-risk genotype (86%; 1C). Hypertension, non-diabetic nephropathy and rapid progression of chronic kidney disease are the relevant clinical factors (100%; 1B), and focal segmental glomerulosclerosis, HIV-associated nephropathy, non-diabetic kidney disease, so-called «hypertensive nephrosclerosis» and end-stage kidney disease are associated with the high-risk genotype (93%; 1B). Not knowing the diagnosis may translate into lack of preparation for rapid progression to kidney failure, inadequate monitoring and the loss of an opportunity to take part in a trial, and identifying the genotype in a patient with kidney disease allows the physician to better estimate the probability of rapid progression and guide monitoring (100%; 1C for both). The panel noted that testing was rare in current practice, performed in few specialised centres and often in a research setting, with no «typical pathway» (100%), and assigned to the nephrologist the responsibility for counselling, ordering the test and returning the result, based on data that primary care physicians feel ill-prepared for this task (100%; 1C) [30].

The two guidance documents diverge in their starting point. The Delphi consensus recommends «considering» the entity, and by implication testing, in every kidney patient of African ancestry, without requiring that the result change management; the KDIGO consensus makes the offer of testing conditional on a cumulative set that includes utility for management and the availability of qualified counselling [1,30]. The Delphi panel did not reach consensus on testing people without kidney disease with a family history of chronic kidney disease or with hypertension, and projected for the time when targeted treatment exists the offer of testing, with counselling, to all patients with suspected disease (100%) and to all Black people or people of sub-Saharan ancestry with chronic kidney disease (79%, with 14% disagreement, the statement with the least support) [30]. The 55 statements are brought together in section 8.7.

5.3. Content of the test and the report

The test is targeted genotyping, not sequencing. The minimum targets are the two polymorphisms that define G1 — c.1024A>G, p.(Ser342Gly), rs73885319, and c.1152T>G, p.(Ile384Met), rs60910145 — and the deletion that defines G2 — c.1164_1169del, p.(Asn388_Tyr389del), rs71785313 —, all on transcript NM_003661.4, and, for reasons developed below, the modifier variant c.792C>A, p.(Asn264Lys), rs73885316 [19,31]. In the West African study, genotyping was performed with custom TaqMan assays, a method suited to known targets and to large scale [7].

The report should state the gene, the transcript with version, the HGVS nomenclature of each variant, the resulting genotype (high risk when G1/G1, G2/G2 or G1/G2; low risk in the other combinations), the status of the modifier variant and, ideally, the phase between G2 and the modifier variant. The ACMG/AMP pathogenicity classification does not apply to these alleles in the usual sense: in ClinVar they are registered as «risk factor», with conflicting classifications when assessed individually, and the modifier variant is classified as benign or likely benign, which reflects precisely the non-Mendelian nature of the locus and should not be read as clinical irrelevance [31].

5.4. Why genotyping without the modifier is insufficient

Inclusion of the p.(Asn264Lys) variant is no longer optional. In cis with a G2 allele, it practically abolishes the risk, and risk prediction based only on the G1 and G2 variants, without testing for the modifier, is considered inadequate [19,20]. The magnitude is large: in G2/G2 genotypes with the variant no case of focal segmental glomerulosclerosis was observed, and in the biobank cohort the presence of the variant in a patient with chronic kidney disease and a high-risk genotype led to the identification of an alternative and potentially treatable cause in 10 of 13 cases (76.9%) [16,19].

There are three limitations to bear in mind. The first is frequency: the variant exists only on the G0 and G2 haplotypes, never on G1, and is present in less than 1% of people of African ancestry, so testing for it resolves only a fraction of cases — about 6% of G1/G2 and G2/G2 carriers in the Million Veteran Program and an allele frequency of 3.2% in a French series [1,33]. The second is phase: the protection demonstrated in vitro exists when the variant is in cis with G2 and not when it is in trans, on the G0 allele, so a report indicating G2 and the variant without phase information leaves the risk indeterminate; determining phase may require parental testing [32]. The third is the specificity of the effect: in low-risk genotypes no independent protection was demonstrated, including in an analysis restricted to G0/G0, so the variant should not be read as a general protective factor against kidney disease [16].

5.5. Histology: what the biopsy adds

In nephrotic forms, biopsy establishes focal segmental glomerulosclerosis or collapsing glomerulopathy, with segmental or global collapse of the capillary tuft and hyperplasia of the overlying epithelial cells, and excludes immune or deposition causes [6,37]. In subnephrotic forms, previously classified as arterionephrosclerosis, biopsy has partial discriminating value. The study by Larsen and colleagues identified three morphological findings associated with two risk alleles — less than 50% of globally sclerotic glomeruli of the obsolescent type, more than one focus of microcystic tubular dilatation and thyroidisation-type tubular atrophy — and tested their combination: the presence of at least two of the three criteria had a sensitivity of 43% and a specificity of 92% in the discovery phase and, in validation by four nephropathologists with experience ranging from less than one year to more than thirty years, a mean sensitivity of 36% and specificity of 86% for the presence of two risk alleles. Reproducibility among observers of very different experience was one of the study's conclusions [3].

The practical reading is twofold. On the one hand, the combination of these findings on a biopsy from a patient of African ancestry with hypertension and chronic kidney disease is a legitimate reason to request genotyping. On the other, the low sensitivity means that their absence does not exclude the entity, and the authors themselves noted that the specificity observed in this clinical scenario may not hold in others [3].

5.6. Biomarkers and prediction of progression

There is no biomarker that measures variant status: serum creatinine with estimated glomerular filtration rate, the urine protein/creatinine or albumin/creatinine ratio and blood pressure define the phenotype and were the outcome measures in the donor cohort, but none provides information on the genotype [6]. Biomarker measurement nevertheless has prognostic potential: plasma concentrations of tumour necrosis factor receptor 1 and receptor 2 and of kidney injury molecule 1 were independently associated with the renal outcome, defined as kidney failure or a sustained 40% decline in estimated glomerular filtration rate, and elevation of all three increases the risk further. Strategies under development include coupling the Kidney Failure Risk Equations to biomarker panels and using genetic or environmental modifiers and polygenic scores as components of these panels [1]. The impact of an independently derived polygenic score on the development of chronic kidney disease in carriers of high-risk alleles has, so far, been modest [20].

5.7. What testing changes in practice: evidence from implementation trials

Two randomised trials assessed return of the result as an intervention, and their results should inform expectations.

GUARDD-US was a pragmatic trial, conducted in routine clinical practice, with recruitment between 1 July 2020 and 30 September 2023, which randomised 6,754 participants with hypertension and self-reported African ancestry, aged 18 to 70 years, without diabetes and without chronic kidney disease or with chronic kidney disease stages 1 to 4, to immediate return of the genetic result or return delayed until the end of the study. The primary analysis, by modified intention to treat, focused on the 954 participants with a high-risk genotype (14.1% of the total), with a mean age of 55.3 years (standard deviation 10.3), 63.8% women, mean systolic blood pressure of 133.1 mmHg (standard deviation 19.2) and 33.4% living below the federal poverty line. The primary outcome, change in systolic pressure at three months, did not differ between groups: a fall of 1.8 mmHg (standard deviation 18.8) with immediate return and 1.5 mmHg (17.3) with delayed return, a difference of −0.3 mmHg (95% CI −2.7 to 2.1; p = 0.78), and an adjusted difference of −0.5 mmHg (95% CI −2.6 to 1.6). There were no differences in diastolic pressure at three and six months or in the trend up to six months [49].

Two secondary results are clinically relevant. In the subgroup with uncontrolled blood pressure at entry there was a significant interaction (p = 0.004, below the Bonferroni threshold of 0.008), with a fall in systolic pressure of 11.4 mmHg (standard deviation 18.7) with immediate return vs 7.1 mmHg (20.3) with delayed return, corresponding to an adjusted mean difference of −4.1 mmHg (95% CI −7.7 to −0.5), and −4.3 mmHg (95% CI −8.0 to −0.5) in those already receiving antihypertensives; in participants with controlled pressure at entry the opposite trend was observed, with a difference of +2.7 mmHg (95% CI −0.2 to 5.6). And immediate return increased screening: at six months, among those without microalbuminuria measured at entry, there were more orders in the immediate return group (95 of 266, 35.7%, vs 43 of 233, 18.5%; difference 17.3%; 95% CI 9.6 to 24.9; p < 0.001), and documentation of a chronic kidney disease diagnosis increased more in that group (31 of 347, 8.9%, vs 10 of 313, 3.2%; difference 5.7%; 95% CI 2.2 to 9.3; p = 0.002), with 24 of 308 (7.8%) vs 1 of 267 (0.4%) among those without chronic kidney disease at entry. There were no differences in the number of antihypertensives prescribed or in self-reported non-adherence [49].

The earlier GUARDD trial, of 2,050 patients of African ancestry with hypertension followed in primary care, showed that patients with a high-risk genotype had higher baseline systolic pressure (137 mmHg, standard deviation 21) than low-risk patients (134, standard deviation 19; p = 0.003) and controls (133, standard deviation 19; p = 0.001), with no significant difference at twelve months; the prespecified outcome of increased urine testing was negative, although in a non-prespecified analysis the increase was greater in high-risk patients, from 17% to 29%, than in low-risk patients and controls [50].

The synthesis of these two trials is that return of the genotype does not by itself lower blood pressure in the general population of hypertensive people of African ancestry, produces a measurable effect in patients with uncontrolled pressure and consistently increases screening and recognition of chronic kidney disease.

5.8. Proposed diagnostic algorithm

5.9. Limits of the tests

Targeted genotyping detects only the alleles tested: a panel that includes G1 and G2 and omits the modifier variant overestimates the risk in a fraction of G2 carriers, and a panel that does not include APOL1 fails to identify the most frequent entity in people of recent African ancestry [19,20]. Exome or genome sequencing identifies the variants, but their interpretation requires targeted haplotype analysis and, for phase, family information [32]. A low-risk result does not exclude genetic kidney disease: in a French cohort, patients with a high-risk genotype generally had no other causal variant, but the reverse also holds, and assessment for monogenic causes remains indicated when the context justifies it [15,20]. Finally, a high-risk genotype in a patient with kidney disease does not prove that the genotype caused the disease, and coexistence with diabetic nephropathy, lupus nephritis, tubulointerstitial nephritis or HIV-associated nephropathy has been documented in the same biopsy series [6,33].

5.10. Prenatal and preimplantation diagnosis

Prenatal diagnosis and preimplantation genetic testing have no application in this entity and are not practised. The reason is conceptual and not technical: this is a common risk allele, of low penetrance, with no recurrence risk in the Mendelian sense, whose presence in the fetus does not allow disease to be predicted [6]. The only area in which the fetal genotype has documented relevance is the risk of pre-eclampsia, where research is active and there is no established clinical application; whether genetic testing can predict pre-eclampsia and improve outcomes has been formulated as a research question, not as practice [45,46].

6. Differential Diagnosis

6.1. Comparative table

6.2. The differentials that matter most in practice

The first real problem is not distinguishing APOL1-associated kidney disease from an exotic entity, but recognising it where a clinical label already exists. Hypertensive nephropathy and nephrosclerosis are the diagnoses that conceal it, and their review in patients of recent African ancestry is the main source of new diagnoses. The elements pointing to APOL1-associated disease are the disproportion between the severity attributed to hypertension and the organ damage, rapid progression under good blood pressure control, younger age — in the histopathological series, mean age 47 vs 56 years — and a biopsy pattern with less obsolescent glomerulosclerosis and more solidified glomerulosclerosis, thyroidisation-type tubular atrophy and microcystic tubular dilatation [3,4,20].

The second is the separation between APOL1-associated disease and monogenic podocytopathy, which are not mutually exclusive but have different consequences. In patients with steroid-resistant nephrotic syndrome or focal segmental glomerulosclerosis, a high-risk genotype generally does not coexist with another causal variant, and families with several affected members may have APOL1-associated disease rather than monogenic disease, which explains familial aggregation without a Mendelian pattern [15]. The distinction matters above all in transplantation: disease mediated by the recipient's high-risk genotype should not recur in the graft, because post-transplant risk depends on the genotype of the donor kidney and not of the recipient, whereas several monogenic forms and immune-mediated focal segmental glomerulosclerosis carry a risk of recurrence [20].

The third is determining which of the findings explains the disease when several coexist. The French series of genotyping in routine practice shows the size of the problem: among patients with a high-risk genotype who underwent biopsy, the diagnoses included HIV-associated nephropathy, primary and secondary focal segmental glomerulosclerosis, class III lupus nephritis, tubulointerstitial nephritis due to sarcoidosis and diabetic nephropathy [33]. The presence of the protective modifier variant in a high-risk patient is, in this context, a strong indicator that the cause lies elsewhere, and should trigger an active search for an alternative aetiology, which was identified in more than three-quarters of cases in a biobank cohort [16].

The fourth is the differential diagnosis in the kidney graft. De novo collapsing focal segmental glomerulosclerosis in a graft may be disease associated with the donor's genotype, triggered by cytomegalovirus or BK virus viraemia, and may coexist with antibody-mediated rejection, as happened in the case of a recipient with a G1/G1 genotype who received a kidney from a donor with a single risk allele [38,51].

7. Treatment and Risk Management Strategies

7.1. General situation: no authorised targeted therapy

There is no approved specific therapy for APOL1-associated kidney disease in any jurisdiction. None of the compounds in development has a Food and Drug Administration (FDA) label, a Drugs@FDA record or a DailyMed entry, nor a European Medicines Agency authorisation, nor an SmPC approved by INFARMED or a package insert registered by ANVISA, because there is no medicine authorised for the indication. Inaxaplin should therefore be described as experimental, and the same applies to AZD2373, MZE829 and the use of baricitinib in this condition [6,22]. Current treatment consists of managing the kidney disease that actually exists, treating the modifiable second hit and making informed decisions in transplant situations [6].

The 2021 Delphi consensus recorded this absence as a statement in its own right, with 100% full agreement, and added that many patients with the subnephrotic form are classified as unspecified chronic kidney disease or as attributed to hypertension, even though intensive blood pressure control does not slow progression. For the time when effective treatment exists, the panel recommended starting the intervention as early as possible, according to trial data (93%), and confirming the diagnosis by kidney biopsy in carriers of a high-risk genotype before treating (86%, with 14% neutral responses); it identified as risks inappropriate treatment and inconsistency of practice (100%), lack of cost coverage by insurers (100%) and classification of the genotype as a pre-existing condition (93%), and argued that the genotype should not increase the cost of health insurance, by legislative means (100%) [30].

7.2. Treatment of the second hit

The intervention with the greatest demonstrated effect is not renal but infectious: in HIV-associated nephropathy, treatment of the infection is the most important modifiable second hit and takes priority over any other measure [6]. HIV testing is mandatory in any compatible nephropathy, for the same reason [6]. Therapeutic exposure to interferon alpha, beta or gamma is a recognised trigger, so prescribing interferon to a carrier of a high-risk genotype with kidney disease requires explicit consideration of the renal risk [1]. In the transplant setting, cytomegalovirus and polyomavirus infections have been implicated as second hits, which reinforces protocol-based virological surveillance [38,48].

7.3. Nephroprotection: what is established and what is not

Management of proteinuric kidney disease follows the general rules, not specific to this entity: blood pressure control, renin-angiotensin system blockade, sodium-glucose cotransporter 2 inhibition, dietary sodium reduction, smoking cessation and correction of the other progression factors. The fundamental caveat, formulated in the KDIGO consensus, is that much of the data on progression of APOL1-associated disease was generated before the widespread use of sodium-glucose cotransporter 2 inhibitors, glucagon-like peptide 1 receptor agonists and dual angiotensin and endothelin receptor antagonists, and that it is not known whether these therapies influence the development or progression of this specific form of kidney disease [1]. Determining the effect of conventional therapies for focal segmental glomerulosclerosis and other glomerulopathies — including calcineurin inhibitors and cytotoxic agents — in carriers of a high-risk genotype has been listed as a research priority, which means that their use in this population is extrapolated and not validated [1].

Since there are no published doses for the condition, the doses below are those of the official labels for the approved indications in the corresponding proteinuric kidney diseases, and their use in APOL1-associated kidney disease is an explicit extrapolation.

7.4. Corticosteroids and immunosuppression

The response to corticosteroids in patients with focal segmental glomerulosclerosis and two risk alleles was similar to that of other patients in the cohort that characterised the phenotype, although progression to end-stage kidney disease was faster [17]. This observation has two consequences: the high-risk genotype is not a reason to deny a trial of corticosteroids when histology and context indicate it, and a response to corticosteroids does not exclude APOL1-associated disease. In severe cases with collapsing glomerulopathy, steroid resistance is frequent: in the Brazilian series of dengue-triggered collapsing glomerulopathy, the first patient, treated with pulsed methylprednisolone followed by prednisone, was steroid-resistant and did not recover kidney function, whereas the second improved spontaneously without immunosuppression, which illustrates the heterogeneity of outcome and the difficulty of attributing the result to treatment [37]. Determining the role of conventional therapies for focal segmental glomerulosclerosis in carriers of a high-risk genotype remains an open question on the research agenda [1].

7.5. Kidney transplantation: decisions about the living donor

The decision about living donation is the only scenario in which the genotype currently changes a concrete clinical decision [6]. The empirical basis is the cohort of North American donors with a median follow-up of 18.5 years, in which 15.3% of Black donors had a high-risk genotype and these had a relative risk of 2.31 (95% CI 1.16–4.61) of reaching an estimated glomerular filtration rate below 45 ml/min/1.73 m², attenuated to 1.91 (95% CI 0.90–4.03) after adjustment for kidney function before donation; the authors concluded that all Black living donor candidates in the United States should be genotyped for better risk stratification [47].

Recommendations differ in strength between jurisdictions. The KDIGO guideline on the evaluation and care of living kidney donors suggests that APOL1 genotyping may be offered to candidates with sub-Saharan African ancestry. The British Transplantation Society recommends offering genotyping, funded by the UK National Health Service, as part of the assessment and risk communication to all potential donors of African heritage under 60 years of age, early in the assessment. The North American practice resource for transplant candidates recommends that counselling and testing be offered to all living donor candidates of sub-Saharan or West African ancestry, regardless of family history, including information on the protective role of the p.(Asn264Lys) variant. In a survey of US transplant centres, half offered the test [1,20].

What to do with the result was the subject of a Delphi panel of 27 participants, of whom 19 (70%) self-identified as Black. The panel reached consensus on four points: candidates should be asked about African ancestry and not about race; testing decisions should be made only after discussion with the donor; disclosure of results to blood relatives and to the recipient should be encouraged but not mandatory; and results should inform shared decision-making and never be used for unilateral decisions by transplant programmes. Equally important is the set of options rejected by more than 77% of the panel: not educating donors about the test or asking their permission (25 of 27 stated they could not accept this option), educating without offering or encouraging the test (22), educating and requiring the test (20), and allowing donors to read results online without the support of a genetic counsellor or physician (23). Prohibiting donation on the basis of genotype was generally rejected. The panel argued that the Black community should be the steward of testing policy, given the risk of racially targeted genetic testing and the history of mistrust, and identified frequent misunderstandings that information should correct: the meaning of recent African ancestry, the relative benignity of a single copy and the fact that, even with two risk alleles, it is more likely not to reach end-stage kidney disease than to reach it [55].

The good practices synthesised by the consensus include a holistic and personalised assessment of benefits and risks for the donor and for the recipient candidate, according to the KDIGO equitable decision framework; individual risk assessment, recognising that the genotype alone may not confer substantive risk; availability of alternatives for the recipient, including another living donor, a deceased donor or dialysis; genetic counselling before and after testing; consideration of available resources; opportunity for follow-up; and potential access to future targeted therapies [1]. The experience of donors themselves shows where the process fails: in interviews with 31 Black people assessed for donation who had the test and agreed to donate, the information received often did not match their expectations or needs, some participants did not recall having had the test, a common concern was that the test would prevent donation, and many considered that candidates should take part in the decision to test and in the use of the results [56].

7.6. Kidney transplantation: deceased donor, graft and recipient

On the graft side, the genotype of the transplanted kidney is the demonstrated determinant: kidneys from deceased donors with a high-risk genotype are associated with faster failure, regardless of donor ethnicity, whereas survival of grafts from donors of African ancestry without risk variants is similar to that of donors without that ancestry. Kidneys from high-risk donors may have pre-existing injury, with podocyte density about 15% lower on pre-implantation biopsy, and lower podocyte density is associated with worse outcomes [1,48].

As for the recipient, the data are mixed. The traditional position was that the recipient genotype did not influence graft survival, and from it derives the practical observation that disease mediated by the recipient's high-risk genotype should not recur in the graft [20]. More recent data suggest that the recipient genotype may also influence post-transplant outcomes, possibly through circulating apolipoprotein L1, a mechanism distinct from the local and podocytic injury of donor variants [1,48]. The case of a Surinamese recipient with a G1/G1 genotype who developed collapsing focal segmental glomerulosclerosis in a graft from a donor with a single risk allele, after antibody-mediated rejection, was presented as an argument for considering routine genotyping in donors of West African ancestry and for not assuming that a single allele in the donor is irrelevant [51].

Organ allocation is an administrative consequence under discussion. Existing prediction tools do not include the graft genotype, although including APOL1 in the Kidney Donor Risk Index improves the performance of the instrument, and considering the genotype instead of race changes the perceived quality of kidneys from African American deceased donors, which may affect their allocation and reduce discard [1]. In the current calculation of the index, Black race of the deceased donor is associated with a 20% increase in the estimated risk of graft loss, and replacing the race variable with the genotype is being studied [48]. The APOLLO network was created to generate the missing prospective data and will assess outcomes of 2,614 deceased donor-recipient pairs, as well as living donor pairs and unpaired deceased donor kidneys, with the participation of the national organ sharing organisation, organ procurement organisations, transplantation and histocompatibility societies and almost all kidney transplant programmes in the United States, and with a community advisory council of Black people with a personal or family history of kidney disease advising from the outset [1,25].

If the donor genotype is validated as a main cause of faster graft failure and of higher risk in living donors, agents effective in native disease should be tested in the transplant setting, with prior demonstration of safety and absence of interaction with immunosuppressants; open questions include whether they should be started immediately after transplantation or donation, to prevent or delay nephropathy, or only when there is evidence of disease, and according to which criteria. The safety of donation by young candidates with a high-risk genotype, as in child-to-parent donations or between young siblings, is a specific point of concern [1].

7.7. Mechanism-targeted therapy: classification and strategies

Drug development in this entity is organised into four strategies, all under clinical investigation and none approved [1,57].

The first is direct blockade of the channel formed by the variant protein, with orally administered small molecules. Inaxaplin, formerly designated VX-147, is the most advanced compound. Its mechanistic basis is direct: in HEK293 human embryonic kidney cells with tetracycline-inducible APOL1 expression, the molecule selectively inhibited channel function, and in a transgenic mouse model homologous to G2 it reduced proteinuria [21]. MZE829 is a second oral compound in phase 2, with separate cohorts for patients with and without concomitant diabetes [58].

The second is reduction of gene expression by subcutaneously administered antisense oligonucleotides. AZD2373 reduces APOL1 expression, and its phase 1 study included, as secondary outcomes, the percentage change in plasma APOL1 protein and the allelic status of participants, which allows the intended pharmacodynamic effect to be verified [1,59].

The third is inhibition of APOL1 induction dependent on the Janus kinase and signal transducer and activator of transcription pathway, a rationale that follows directly from the role of interferon as a second hit. The JUSTICE trial tested baricitinib, an oral inhibitor of this pathway, in patients with a high-risk genotype and focal segmental glomerulosclerosis or chronic kidney disease attributed to hypertension [41,57].

The fourth consists of downstream anti-inflammatory approaches, targeting mediators activated by the variant protein, still at a preclinical or exploratory stage [57]. Development is oriented towards focal segmental glomerulosclerosis, HIV-associated nephropathy, COVID-19-associated nephropathy, collapsing glomerulopathy and solidified glomerulosclerosis; the effect in carriers of a high-risk genotype with diabetic nephropathy — an important combination given the high frequency of diabetes in populations with recent African ancestry — or with membranous nephropathy and other forms of chronic kidney disease remains to be explored [1]. To date there is no clinical programme of genome editing or gene therapy for this entity: the possibility of using gene editing, namely by clustered regularly interspaced short palindromic repeats, has been formulated as a research question alongside targeted therapies, and not as a strategy under trial [1].

7.8. Doses and regimens used in trials

The doses that follow are clinical trial doses, of drugs not authorised for this or any other indication, and do not constitute recommended dosing.

7.9. Clinical trials: design and results

7.10. Limits to the interpretation of the trials

The available data do not allow indirect comparison between compounds, and no therapeutic trial has so far reported a hard clinical outcome in this entity [6]. The phase 2a trial is the most cited evidence and is, strictly speaking, a single-arm, open-label study, with thirteen participants assessed on a three-month surrogate outcome and no control group, so the 47.6% reduction in proteinuria is encouraging but does not demonstrate preservation of kidney function [6,21]. The phase 2b study has the same uncontrolled design, and in it the magnitude of the effect depended on the population: the reduction was substantial in the cohort with modest proteinuria and, in the cohort with type 2 diabetes, the confidence intervals crossed zero, for both albuminuria and proteinuria, which is compatible with no effect in that population; the intervals were, moreover, described as post hoc and the results exist only in a sponsor press release [22]. Relative reductions should be read alongside absolute values: in cohort 1, the mean baseline albumin/creatinine ratio was 0.28 g/g and, in cohort 2, 0.67 g/g, modest values compared with the protein/creatinine ratio of 0.7 to 10 g/g required in the phase 2a trial [21,22].

Negative trials and execution problems should be included in the balance. JUSTICE, which tested Janus kinase pathway inhibition, closed in April 2026 with only 8 participants enrolled in a placebo-controlled randomised design, and did not deposit results: a trial of this size has no power to assess the proposed outcome [41]. In the inaxaplin phase 1 studies there were cohort interruptions by sponsor decision, one owing to electrocardiogram equipment failure and another owing to COVID-19 exposure, with 21 participants discontinuing for these reasons, which limits the interpretation of tolerability at the highest doses [60].

The exclusion criteria delimit applicability. Diabetes mellitus, sickle cell disease, transplantation and uncontrolled hypertension are excluded from the pivotal trial, so any efficacy demonstrated will not automatically extend to these populations, which are precisely those in which APOL1-associated kidney disease most often coexists with other causes [13,40].

There is also a concern specific to this entity, arising from the physiological function of the protein. It is not known whether blocking or reducing APOL1 affects the risk of trypanosomiasis; although rates in sub-Saharan Africa are low, with fewer than a thousand cases per year, weighing the prevention of kidney failure against increasing the risk of infection is necessary, and surveillance for trypanosomal and other infections should be part of trial design. Inclusion of African centres in randomised trials has been identified as a priority, because studies in admixed African American populations may not represent outcomes in continental Africa, where the prevalence of the variants and the rates of sickle cell disease and HIV infection differ, and where post-marketing availability will be decisive [1].

7.11. Regulatory status

7.12. Developments in the last twelve months, with level of evidence

7.13. Agents to avoid and precautions

There is no established list of drugs contraindicated by genotype. Precautions derive from the known mechanisms and from the labels of the drugs used. Exposure to interferon alpha, beta or gamma is the trigger with the strongest evidence for glomerular disease in carriers of a high-risk genotype, so prescribing interferon to these patients requires explicit consideration and monitoring of proteinuria [1]. In symptomatic treatment, sparsentan is contraindicated in pregnancy and in co-administration with angiotensin receptor blockers, endothelin receptor antagonists or aliskiren, requires discontinuation of renin-angiotensin-aldosterone system inhibitors before starting, and carries a risk of hepatotoxicity with mandatory three-monthly monitoring of aminotransferases and bilirubin, as well as hypotension, acute kidney injury, hyperkalaemia and fluid retention [54]. Dapagliflozin requires assessment of kidney function before starting, prior correction of volume depletion and withholding at least three days before procedures with prolonged fasting [53]. Angiotensin receptor blockers require a reduced starting dose in patients with possible intravascular depletion and are not recommended below 6 years of age or with an estimated glomerular filtration rate below 30 ml/min/1.73 m² in the paediatric population [52].

7.14. Pregnancy, lactation, perioperative care and transition

In pregnancy, the entity has two distinct implications. The first is maternal kidney disease, which follows the general rules for chronic kidney disease in pregnancy, noting that sparsentan is contraindicated and requires exclusion of pregnancy before starting, effective contraception during treatment and for two weeks afterwards, and information for prepubertal adolescent girls and their guardians on the need for contraception when they reach reproductive potential [54]. The second is the risk of pre-eclampsia associated with the fetal genotype, with an odds ratio of about 1.8 to 1.9 and a population attributable risk of 12.3% in Black women, which does not currently change obstetric surveillance or create an indication for testing, but is the subject of active research on assessment protocols [45,46]. Long-term pregnancy outcomes in living donors with a high-risk genotype have been listed as a knowledge gap [1].

No data specific to this entity on lactation were identified in the sources consulted. Perioperatively, the relevant precautions are those of chronic kidney disease and of the drugs in use, namely withholding dapagliflozin before procedures with prolonged fasting [53]. In the transition to adult care, the issue specific to this entity is continuity of surveillance in a patient who may have been genotyped in childhood for a glomerulopathy and who remains at risk of progression, and the need to repeat the genetic information in age-appropriate terms, given that the documented misunderstandings about the meaning of a single copy and about the real probability of end-stage kidney disease are frequent even in informed adults [1,55].

8. Reproductive Options and Family Counselling

8.1. What does not apply

Counselling for this entity begins with an important negative. An APOL1 result is not a diagnosis, does not establish the cause of an individual's kidney disease, does not predict that a healthy carrier will develop kidney failure and carries no recurrence risk to counsel on in the Mendelian sense [6]. Consequently, prenatal diagnosis and preimplantation genetic testing have no application and are not practised: these are common alleles, of low penetrance, whose presence in the fetus does not allow disease to be predicted [6]. The reproductive counselling that makes sense is that of maternal kidney disease, when present, and not that of the genotype.

8.2. Risks of transmission and of disease

Transmission of the alleles follows simple Mendelian segregation, but the risk of disease does not. Each parent transmits one of their two alleles at the locus, so the probability of a child inheriting a high-risk genotype depends on the genotypes of both parents: if both are risk homozygotes, all children will have a high-risk genotype; if both have one risk allele, a quarter of the children will have a high-risk genotype; and if only one parent has risk alleles, no child will have a high-risk genotype unless the other also transmits a risk allele. The point that counselling must convey is that inheriting a high-risk genotype is not equivalent to developing disease: about 15% of carriers will develop kidney failure and 5% to 8% will develop focal segmental glomerulosclerosis, and the estimated lifetime risk of focal segmental glomerulosclerosis with two alleles is about 4% in the absence of a known second hit [1,16,17].

The risk attributed to a single copy has changed status and counselling should reflect this change, without exaggerating it. For more than a decade the carrier of one allele was considered essentially unaffected, and this remains the central message in North American clinical settings [6,55]. West African data have, however, demonstrated measurable risk with a single allele, with odds ratios of 1.18 for chronic kidney disease and 1.61 for focal segmental glomerulosclerosis, small magnitudes in absolute terms but relevant in a population in which 43% of individuals are monoallelic [7]. The honest formulation is that one copy confers a slightly increased risk, not a high individual risk.

Information about the protective modifier belongs in counselling whenever it is available. In genotypes with G2, the presence of p.(Asn264Lys) in cis reduces the risk to levels comparable to those of non-risk genotypes, and its absence maintains the risk; in genotypes with G1 the variant does not exist, so there is no such information to give [19,20]. When phase is unknown, counselling should state that the interpretation is indeterminate and that parental testing may resolve it [32].

8.3. Cascade testing and family disclosure

The rules of disclosure are the general ones for genetic results: the process involves providing support and resources to the patient to pass the information on to the family, if they wish, and relatives may then seek counselling and testing [1]. Having a relative with a high-risk genotype is in itself one of the clinical conditions that legitimise testing an individual, alongside having kidney disease and being a living donation candidate, but only when the other cumulative conditions are met, including the possibility that the result will change management [1]. In transplant practice, the Delphi panel consensus was that disclosure of results to blood relatives and to the recipient should be encouraged, but not mandatory [55].

Three additional considerations should be anticipated in counselling. The first is the risk of stigmatisation associated with a test that targets, by definition, people of a particular ancestry, in a field historically marked by confusion between race, ethnicity and ancestry; the correct formulation, established by consensus, is to ask about African ancestry and not about race, and to remember that status is determined by genotyping and not by appearance or self-identification [1,6,55]. The second is the implications for genetic relatives, who did not ask for the test. The third is the possibility of unexpected revelations about paternity, inherent in any family study [1].

8.4. Counselling of minors

Testing of an asymptomatic minor follows the general rules of clinical genetics: it is justified when the result may modify surveillance, guide management or allow treatment from childhood. In this entity, the consensus considers that, in children with focal segmental glomerulosclerosis and chronic kidney disease, genotyping may inform prognosis, and that early targeted testing in infants and children would allow early surveillance and intervention, but that unnecessary testing is costly and potentially anxiety-provoking; the formal recommendation for screening and early testing in childhood is expressly deferred until effective treatments exist [1]. The practical implication is that, at present, testing a minor is justified when there is kidney disease to characterise or when the child is the donor candidate — a situation in which the safety of donation by young candidates with a high-risk genotype is flagged as a specific concern — and not as screening for future risk [1].

8.5. Psychological impact and patient information

The effect of learning the result has been measured. In a prospective cohort of 200 Black participants followed for three months at a centre in the US Midwest, in which 17.5% (35 of 200) had a high-risk genotype, worry about kidney problems was frequent at baseline: 39% were extremely worried, 34% very worried, 12% somewhat worried, 10% a little and 6% not at all worried, with higher levels in women and in those with a family history of kidney disease and hypertension. At three months, worry remained unchanged in 45% of participants, decreased in 36% and increased in 19%, and high-risk genotypes were more frequent among those whose worry increased than among those in whom it decreased (29% vs 11%; p = 0.02). Worry specifically related to having a high-risk genotype fell from 14% at baseline to 8% at three months [63]. The reading is that the result produces a measurable but limited and transient increase in worry, which supports the requirement for qualified counselling without justifying refusal of the test when it is indicated.

The quality of information is a documented problem and not a matter of form. In interviews with 31 Black people assessed for living kidney donation who had the test and agreed to donate, the information received often did not match their expectations or needs and some participants did not recall having had the test [56]. Among the misunderstandings that information should correct are the meaning of recent African ancestry — which refers to the last few thousand years —, the relative benignity of a single copy, and the fact that, even with two risk alleles, it is more likely not to reach end-stage kidney disease than to reach it [55]. Among the operational requirements of a testing programme, the consensus includes co-designed educational material adapted to literacy level, counselling before and after testing and involvement of community leaders, experts and religious leaders in high-prevalence populations, as well as authentic community involvement in the design and communication of studies, so as to avoid racialisation of the test [1].

The wish to know the genotype, even without treatment, is documented in the community to which the test is directed. In community deliberations with participants who all identified as Black, 95% supported offering the test to patients of African ancestry with kidney disease, knowing there was no treatment: 64% favoured offering it to all Black patients, 21% only to those with risk factors for chronic kidney disease and 10% only to those with signs of chronic kidney disease [30]. The 2021 Delphi consensus systematised the content of pre- and post-test counselling, the advantages and disadvantages to discuss and the need to train nephrology trainees, recognising that there is no established good practice and that it requires research; the full set is in section 8.7 [30].

8.6. When the result may inform reproductive decisions

The consensus accepts, among the justifications for testing, that the result may relieve significant anxiety or inform reproductive decisions, instead of or in addition to strict clinical utility [1]. The content of this reproductive information is, however, limited and should be presented as such: it allows a couple to know whether their children may inherit a high-risk genotype and, consequently, a modest risk dependent on future exposures, not a deterministic risk of disease. The areas in which the information could carry greater weight — the relationship between fetal genotype, maternal genotype and pre-eclampsia, and appropriate assessment protocols — remain under investigation and do not support clinical decisions [45,46].

8.7. The 55 statements of the 2021 Delphi consensus

The 2021 Delphi consensus grouped the statements into three domains: counselling, genotyping and diagnosis (A1–A24); disease knowledge and education (B1–B18); and vision for a future with treatment (C1–C13). Agreement is the proportion of voters who answered «agree» or «strongly agree», with the consensus threshold set at 75%. On the GRADE scale, 1B corresponds to a strong recommendation with moderate-quality evidence and 1C to a strong recommendation with low- or very-low-quality evidence; several statements rely only on expert opinion or on indirect evidence, for example from living donors or hypertensive patients. The authors themselves emphasised that the statements are not exhaustive, do not constitute clinical practice guidelines and reflect a panel with few specialists per area [30].

In clinical reading, statements A15 to A21 work as a counselling checklist: before testing, the patient should know that most carriers do not develop chronic kidney disease, that there is no specific treatment and what the implications are for insurance and for relatives; after testing, the conversation differs between high and low risk mainly in the explanation of the absence of targeted treatment and in the invitation to trials, with reduction of anxiety and information on modifiable factors maintained in both [30]. Concerns about insurance and discrimination reflect the North American context and should be translated into the legal framework of each country.

9. Special Considerations: Population Diversity and Epidemiology

9.1. What each figure measures

The epidemiology of this entity requires a distinction that is rarely made in summary texts: the frequency of the risk genotype is not the prevalence of the disease. The curation record used in this work itself declines to record a disease prevalence, on the grounds that multiplying the frequency of a genotype by a lifetime risk would produce a cumulative incidence and not a point prevalence, and would silently assume a fixed exposure to the second hit [6]. Orphanet, for its part, records an «unknown» prevalence for the only entity to which APOL1 is mapped, hereditary steroid-resistant nephrotic syndrome, whose value does not apply to the risk allele [6,9]. The figures that follow are therefore presented for what they actually measure: genotype frequencies in populations, attributable proportions of cases in clinical series and relative risks in cohorts.

9.2. West Africa: the ancestral region

The highest frequencies of the high-risk genotype are found in West Africa. In the H3Africa network study, with 8,355 participants from Ghana and Nigeria, 43.0% were monoallelic and 29.7% biallelic, that is, almost a third of the population studied had a high-risk genotype; prevalence was higher among participants with chronic kidney disease than among controls (31.6% vs 25.7%). This frequency is higher than the 11.0% to 16.5% described in African Americans, an expected difference because African Americans have about 20% European admixture, and higher than the 7% to 11% of East Africa. Allele frequencies in sub-Saharan Africa range from 0% to 45% for G1 and from 0% to 24% for G2 [7].

Sub-regional estimates, brought together in the KDIGO consensus, reach up to 50% high-risk genotypes in south-eastern Nigeria, 37% in central Ghana and 25% in south-western Nigeria. Data on the other sub-Saharan countries are scarce, although the few available studies show significant prevalences of the variants throughout the region. Population size translates these frequencies into potential disease burden: more than one hundred million people in sub-Saharan Africa may have two high-risk alleles, which the consensus describes as a catastrophically high potential burden of APOL1-associated kidney disease in that region [1].

One observation from the West African study deserves attention because it runs counter to intuition: the odds ratio for focal segmental glomerulosclerosis associated with two risk alleles was about 2, much lower than the ratio of about 10 described in African Americans. The explanations proposed by the authors are survival bias in the African American population and the possibility that European admixture has altered the genetic background in which the phenotype is expressed [7]. In public health terms, a 25% increase in the odds of chronic kidney disease and an 84% increase in the odds of focal segmental glomerulosclerosis represent, in a population of this size and with high and early mortality from chronic kidney disease, a substantial burden [7].

9.3. East and Southern Africa

In East Africa frequencies are lower, between 7% and 11% high-risk genotypes [7]. The allelic pattern differs: G2 has similar frequencies in all African populations studied, between 3% and 8%, whereas G1 is common only in the Yoruba of West Africa, where it reaches 39% [5]. In South Africa, the population allele frequency was 7.3% for G1 and 11.1% for G2, with 2% of controls having a high-risk genotype, vs 79% of patients with HIV-associated nephropathy [36]. In the Democratic Republic of the Congo, 7.0% of children from the general population had a high-risk genotype, 20.4% were G1/G0 and 14.8% G2/G0 [18].

9.4. United States

In the United States, estimated allele frequencies among African Americans are 20% to 22% for G1 and 13% to 15% for G2, with 10% to 15% of the population having a high-risk genotype; the GeneReviews note uses the value of 13% and describes a 3- to 30-fold increase in the risk of various forms of kidney disease [1,10]. Among African Americans, about 15% to 20% of carriers of two alleles develop chronic kidney disease, and 40% to 60% of patients with primary focal segmental glomerulosclerosis have two risk alleles [1]. The proportion of cases explained is 18% for focal segmental glomerulosclerosis and 35% for HIV-associated nephropathy, and it is estimated that eliminating the effect would reduce the incidence of both by 67% [17].

In hospital series, the frequency of the high-risk genotype varies very markedly with phenotype: in the cohort of 107,696 individuals from two biobanks, it was 9.2% in patients with focal segmental glomerulosclerosis of all ancestries, 1.8% in patients with chronic kidney disease without focal segmental glomerulosclerosis and 0.8% in controls; among high-risk patients with focal segmental glomerulosclerosis, 85.3% were of African ancestry and 14.4% of mixed ancestry [16]. In the GUARDD-US trial, among 6,754 hypertensive adults of self-reported African ancestry, 14.1% had a high-risk genotype [49]. Among living kidney donors, 15.3% of Black donors had a high-risk genotype [47]. In a prospective genotyping cohort at a Midwest centre, the frequency was 17.5% [63]. According to the 2021 Delphi consensus, end-stage kidney disease affects about 750,000 people in the United States, with an incidence about three times higher in Black people; with a chronic kidney disease risk of 15% to 20% in carriers of two alleles and 10% to 15% of the Black population with a high-risk genotype, up to 3% of the Black population will be affected, and high-risk genotypes are found in 16% to 23% of Black people with non-diabetic chronic kidney disease, which shows that the genotype does not explain all the excess incidence; social determinants, racism, residential segregation, medical mistrust and lack or inadequacy of health insurance contribute to the onset and progression of the disease [30].

9.5. Europe

European data are scarce and come mainly from France. In a cohort of 152 patients from 139 families with steroid-resistant nephrotic syndrome or focal segmental glomerulosclerosis, of African or French Antillean origin, the high-risk genotype was present in 43.1% vs 18.9% in a control population (p < 0.0001), and was more frequent in patients originating from the Antilles than from Africa; it was more frequent in young adults, but also occurred in children, and was associated with a poor renal prognosis. In living donors of African ancestry in France, carriers of a low-risk genotype had a smaller increase in estimated glomerular filtration rate after donation and a smaller baseline kidney volume than Caucasian donors, an observation suggesting that the genotype does not capture all the risk associated with ancestry [15]. In the United Kingdom, the UK Biobank analysis brought together in the consensus compared 2,853 G0/G0 participants, taken as reference, with 2,273 G0/G1, 1,219 G0/G2, 644 G1/G1, 320 G1/G2 and 153 G2/G2, and found odds ratios for albuminuria or estimated glomerular filtration rate below 60 ml/min/1.73 m² of 1.0 (95% CI 0.8–1.3) for G0/G1, 1.1 (0.9–1.4) for G0/G2, 1.4 (1.1–1.9) for G1/G1, 1.6 (1.1–2.2) for G1/G2 and 1.2 (0.7–2.0) for G2/G2 [1].

The European gaps are explicitly acknowledged: the consensus identifies as areas of need prevalence data in Black populations of the United Kingdom, Belgium and France, as well as the Caribbean, Brazil, sub-Saharan Africa and any populations originating from the diaspora [1]. No prevalence data on APOL1 risk genotypes in Portugal, nor Portuguese studies on the entity, were identified in the sources consulted; the absence of data in the sources consulted does not demonstrate the absence of published data, and the gap is recorded as such.

9.6. Brazil

Brazil is, owing to its population composition, one of the regions with the most data outside the United States and Africa, and the results illustrate the effect of genetic admixture. In the study from the south of the country, with 106 haemodialysis patients of African ancestry and 106 healthy first-degree relatives as controls, the high-risk genotype was ten times more frequent in patients (9.4% vs 0.9%; odds ratio 10.95; standard error 1.49; p = 0.0017), whereas the frequency of genotypes with one allele did not differ between groups. In an expanded series of 274 haemodialysis patients, carriers of two alleles started dialysis 12 years earlier than those without risk alleles (coefficient −12.3; 95% CI −18.1 to −6.5; p < 0.001), a difference that remained at 9 years after adjustment for sex, smoking, body mass index and diabetes (−8.97; 95% CI −14.39 to −3.56); a single allele showed only a non-significant trend (−2.96; p = 0.255). Allele frequencies in controls were 6% for G1 and 1.4% for G2, lower than in North America, but the effect of two alleles was strong and the phenotype particularly aggressive. Using first-degree relatives as controls tends to underestimate the association, because these relatives are more likely to carry risk alleles than individuals drawn at random from the population [12].

The integrative review of the Brazilian literature, based on searches of PubMed, Lilacs and SciELO between 2010 and October 2025, describes regional variation in allele frequencies between 10% and 22%, with values up to 22% in the north-east and, in chronic kidney disease cohorts from the south, 15.3% for G1 and 5.9% for G2, levels comparable to those of African American, Caribbean and sub-Saharan populations. Studies from the south-east associated high-risk genotypes with paediatric nephrotic syndrome with accelerated progression, even with a single allele, with high-risk genotypes present in about 8% of paediatric cases. In patients with lupus nephritis from the north-east and south-east, frequencies were similar to those of controls, with a low prevalence of high-risk genotypes, which differs from African American cohorts. In Brazilian idiopathic collapsing glomerulopathy, 58.6% of patients had causal genetic variants, of which 80.5% were high-risk APOL1 genotypes and 19.5% Mendelian variants. In terms of national context, the Brazilian dialysis registry between 2000 and 2012 indicated 40.5% of patients self-identified as of African descent, whereas 56.2% of the total population self-declared as Black or brown (preta or parda) in the national household sample survey between 2012 and 2019, and the mean national hospital mortality from kidney failure in the last decade was 13.29 per 100,000 inhabitants, ranging from 12.08 to 14.59 [14].

The Bahia study adds baseline population data: among 889 newborns from a population survey, the frequency of the high-risk genotype was 1.2%, vs 4.7% in 424 patients with biopsy-proven kidney disease (p < 0.001), whereas sickle cell trait had an identical frequency in the two groups (4.7%). Progression to end-stage kidney disease increased in a graded manner with the number of risk alleles, with 10%, 14% and 56% for zero, one and two alleles (p < 0.001) [13]. In the transplant setting, among Brazilian haemodialysis patients 9.2% had two risk variants and 13.2% one variant, vs 0.9% and 13.2% in controls, and in a Brazilian study of paediatric transplant recipients 8.4% had two variants; genetic testing is not available in Brazilian transplant centres, being restricted to research protocols at some universities and at very high cost [48].

9.7. Caribbean and the rest of the diaspora

The data are fragmentary. The French observation that the high-risk genotype was more frequent in patients originating from the French Antilles than from Africa suggests high frequencies in Caribbean populations, and the consensus explicitly lists the Caribbean among the regions where prevalence data are lacking [1,15]. The global mapping brought together in the consensus includes estimates for Caribbean and Central and South American countries, with patterns of high G1 and G2 frequency similar to those described in populations of African descent on the American continent [1,14].

9.8. Special populations and equity issues

APOL1-associated kidney disease disproportionately affects communities with limited resources and infrastructure for research and care delivery, so access is part of the clinical problem and not an accessory. When treatments exist, access to nephrologists with experience of this entity will be decisive, as will the affordability of new therapies; the priority of including African centres in randomised trials stems from the same concern, reinforced by the finding that results obtained in admixed African American populations may not transfer to continental Africa, where allele frequencies, rates of sickle cell disease and HIV infection and the availability of care differ [1].

The equity dimension also has a protective side. APOL1 testing has the potential to reinforce racialised medicine, given the persistent confusion between race, ethnicity and ancestry, and the Delphi panel on testing policy, with 70% of participants self-identified as Black, argued that the Black community should be the steward of this policy, rejecting both the absence of information and consent and the imposition of testing, and rejecting the prohibition of donation on the basis of genotype [1,55]. Protection against stigmatisation and discrimination, including the existence of policies and legislation on the use of genetic results, was listed as a priority research question, together with the long-term outcomes of living donors with a high-risk genotype [1].

The 2021 Delphi consensus devoted 18 statements to knowledge of the entity and education. The panel agreed that awareness of kidney disparities in Black people varies among stakeholders (100%) and that the lack of Black representation among them aggravates these disparities (100%); that many nephrologists know the entity, but that knowledge is inadequate among regulators, payers, policymakers and non-specialist professionals (100%), and low in the Black community (93%), with consequences for participation in research and trials (86%) and for kidney recipients' and donors' lack of awareness of their own status (100%). Absence of treatment, doubts about reimbursement of the test and fear of discrimination in life insurance, health insurance and employment are barriers to patient education (100%); lack of education reinforces mistrust of the medical establishment (79%, with 14% disagreement), rooted in historical and current mistreatment of Black people by medical and other institutions, and this mistrust may be a barrier to testing (100%). For education, the panel assigned a role to nephrologists, primary care physicians, nurses, community leaders and patient networks; required educators with up-to-date knowledge, training in genetic risk communication and in cultural and structural competence, able to use lay language and to convey ancestry-based risk with sensitivity; recommended clear and accurate materials, ideally with Black people, referral to reliable sources such as medical societies, patient associations and health systems, and education delivered with community leaders and groups in settings that are familiar and comfortable for patients (100% for all). It also recorded that use of the genotype instead of race in the Kidney Donor Risk Index was under study (93%) [30].

9.9. Reference networks and resources

The relevant European reference network is ERKNet, the European reference network for rare kidney diseases, one of the 24 networks approved by the board of Member States and funded by the EU4Health programme for 2021 to 2027, which covers among its thematic areas hereditary glomerulopathies and immune glomerulopathies, as well as tubulopathies, thrombotic microangiopathies, congenital malformations and ciliopathies, paediatric chronic kidney disease, paediatric dialysis and transplantation and molecular diagnosis; it provides an expert finder, a clinical patient management system for virtual consultation, guidelines and clinical pathways, the ERKReg registry, postgraduate training and a European patient advocacy group [23].

In the United States, NephCure is the patient organisation dedicated to rare proteinuric kidney diseases, with patient navigation services, an online community, support groups, educational material, information on clinical trials and a specialist directory; it covers focal segmental glomerulosclerosis, nephrotic syndrome and immunoglobulin A nephropathy, and APOL1-associated kidney disease appeared, at the time of consultation, mainly in connection with a patient-focused drug development meeting, without dedicated informational content on the home page [24]. Among the structures specific to this entity, the community advisory council of the APOLLO network stands out, made up of Black people with a personal or family history of kidney disease, which has advised the investigators since the start of the project, and whose model of community engagement is presented by the consensus as an example for the design and communication of studies in this area [1,25]. Patient associations represented on the clinical practice consensus panel included the American Association of Kidney Patients and the Texas Kidney Foundation [30].

10. Illustrative Clinical Case

10.1. Dengue-triggered collapsing glomerulopathy in a 61-year-old man

The case chosen is the second patient of a Brazilian series of three cases of collapsing glomerulopathy triggered by dengue virus infection in carriers of a high-risk APOL1 genotype, described as the first published association between dengue and this form of glomerular disease [37].

A 61-year-old man, with a previous diagnosis of arterial hypertension, was referred on the seventh day after the diagnosis of dengue, confirmed by a positive NS1 antigen test, because of laboratory abnormalities. On admission he was prostrate, with a diffuse rash, mild dehydration, blood pressure of 109/75 mmHg and no oedema. Tests showed blood urea nitrogen of 75 mg/dl and creatinine of 4.6 mg/dl, corresponding to an estimated glomerular filtration rate of 14 ml/min/1.73 m², when the previous value had been 98 ml/min/1.73 m²; there were no acid-base or electrolyte disturbances; haemoglobin was 13.9 g/dl and platelets 99,000/mm³. Urinalysis revealed haematuria, with proteinuria of 2.64 g in 24 hours. Total cholesterol was 140 mg/dl, low-density lipoproteins 75 mg/dl, high-density lipoproteins 40 mg/dl and serum albumin 3.0 g/dl; complement fractions C3 and C4 were normal, and serologies for human immunodeficiency virus, hepatitis B and C, cytomegalovirus and Epstein-Barr virus, as well as the autoimmunity work-up, were negative [37].

A kidney biopsy was performed. Light microscopy identified 3 of 28 glomeruli with segmental synechiae and 5 with collapse of the capillary loops associated with podocyte hypertrophy and hyperplasia, as well as acute tubular necrosis. Immunofluorescence revealed IgM and C3 with an intensity of 1+ out of 3+ in the sclerotic areas. Electron microscopy showed foot process effacement in 60% of the capillary loops, without tubuloreticular inclusions. The overall picture was considered compatible with collapsing glomerulopathy [37].

No immunosuppression was prescribed. Kidney function improved progressively and proteinuria decreased spontaneously. At the last assessment, creatinine was 1.2 mg/dl and proteinuria 0.4 g in 24 hours, on an angiotensin II receptor blocker. APOL1 genotyping, performed after the episode, identified a high-risk G1/G1 genotype [37].

Figure 1 of the article, which was consulted, shows the characteristic histological findings: panel (a) documents segmental glomerular collapse with podocyte hypertrophy and hyperplasia, in periodic acid-Schiff staining at 400-fold magnification, and panel (b) shows collapse of the glomerular tuft with the same podocyte changes in methenamine staining — https://pmc.ncbi.nlm.nih.gov/articles/PMC12640018/figure/fig1/ [37].

10.2. Contrast with the first patient of the same series

The first patient of the series illustrates the opposite extreme of outcome and deserves mention for the contrast. She was a 54-year-old Black woman with hypertension, diabetes and stage 4A2 chronic kidney disease, with a baseline estimated glomerular filtration rate of 22 ml/min/1.73 m² and albuminuria of 230 mg per day, admitted four days after the diagnosis of dengue with fatigue, drowsiness and facial palsy, and with a right frontoparietal ischaemic stroke documented by magnetic resonance imaging. On admission, creatinine was 11 mg/dl, with an estimated glomerular filtration rate of 4 ml/min/1.73 m², and 24-hour proteinuria was 24.4 g, with albumin of 1.9 g/dl. She started haemodialysis and received methylprednisolone 1 g per day for three days, followed by prednisone 80 mg per day. The biopsy, performed on day 25 of admission, showed 14 of 18 glomeruli globally sclerotic, 2 with segmental synechiae and only 2 viable, these with collapse of the capillary loops and podocyte hypertrophy and hyperplasia, accompanied by acute tubular necrosis and interstitial fibrosis with tubular atrophy in 60% of the sample; immunofluorescence revealed C3 deposits with an intensity of 2+ out of 3+ in the sclerotic regions, and electron microscopy was not performed owing to the absence of viable glomeruli in the remaining tissue. The patient was steroid-resistant, did not recover kidney function and remained on replacement therapy. Subsequent genotyping revealed a G1/G1 genotype [37].

10.3. Interpretation of the cases

Three points about these cases deserve comment, and two of them are caveats.

The first is that the same genotype, G1/G1 in both patients, produced opposite outcomes: almost complete recovery in one, end-stage kidney disease in the other. The difference is associated with functional reserve before the episode — previous estimated glomerular filtration rate of 98 vs 22 ml/min/1.73 m² — and with the degree of chronicity on biopsy, with 60% interstitial fibrosis and tubular atrophy and only two viable glomeruli in the patient who did not recover. This observation is consistent with the nature of the genotype as a risk factor and not as a determinant of outcome [1,37].

The second is that the management described in these reports should not be read as routine. The patient who recovered received no immunosuppression and improved spontaneously, whereas the patient who did not recover received high-dose corticosteroids without benefit. In a case series design, without a comparison group, it is not possible to attribute the outcome to treatment or to its absence, and the role of conventional therapies for focal segmental glomerulosclerosis in carriers of a high-risk genotype remains an explicitly open research question [1,37]. The response to corticosteroids in this entity is not, in general, worse than in other patients with focal segmental glomerulosclerosis, although progression to end-stage kidney disease is faster [17].

The third is the diagnostic lesson. In both cases genotyping was performed after the acute episode, which is the usual sequence in practice and means that the genetic information did not guide the initial clinical decision. In both, the investigation excluded the classic infectious and autoimmune causes before attributing the lesion to the genotype-infection combination, a procedure that corresponds to the requirement not to use the genotype as a diagnostic shortcut [16,37]. Dengue thus joins the human immunodeficiency virus and SARS-CoV-2 on the list of infections capable of triggering collapsing glomerulopathy in carriers of a high-risk genotype, with the particularity of being endemic in regions where these genotypes are frequent [14,29,37].

11. Prognosis and Long-Term Follow-up

11.1. Prognosis with data

The prognosis of this entity must be stated on two levels, because the prognosis of an asymptomatic carrier and that of a patient with established nephropathy are radically different.

For the asymptomatic carrier of a high-risk genotype, the prognosis is dominated by low penetrance: about 15% will develop kidney failure and 5% to 8% will develop focal segmental glomerulosclerosis, with an estimated lifetime risk of focal segmental glomerulosclerosis of about 4% in the absence of a known second hit. Most carriers will never have overt kidney disease, and the formulation that the testing policy panel considered essential to convey is that, even with two risk alleles, it is more likely not to reach end-stage kidney disease than to reach it [1,16,17,55]. The prognosis changes profoundly in the presence of a second hit: in untreated HIV infection, the risk of HIV-associated nephropathy in a high-risk carrier approaches 50% [17].

For the patient with established kidney disease, the high-risk genotype is a consistent marker of faster progression. In the AASK trial, in patients with chronic kidney disease attributed to hypertension, the composite outcome of end-stage kidney disease or doubling of creatinine occurred in 58.1% of carriers of two alleles vs 36.6% of the others, with a hazard ratio of 1.88 (p < 0.001), and progression of chronic kidney disease had an odds ratio of 1.88 (95% CI 1.46–2.42); in the CRIC study, high-risk Black patients had a faster decline in estimated glomerular filtration rate and a higher risk of the composite renal outcome than White patients, with and without diabetes [1,28]. In African Americans of the ARIC cohort, two risk alleles were associated with an odds ratio of 1.92 (95% CI 1.19–3.10) for incident kidney failure, 1.51 (1.01–2.24) for incident chronic kidney disease and 2.43 (1.00–5.95) for progression; in the Jackson Heart Study, G0/G2 carriers had an odds ratio of 9.05 (95% CI 1.79–45.85) for incident kidney failure, with a very wide interval, 1.65 (1.06–2.57) for incident chronic kidney disease and 1.88 (1.04–3.40) for incident albuminuria [1].

The age at which events occur is earlier. Focal segmental glomerulosclerosis with two alleles begins earlier and progresses faster to end-stage kidney disease, with similar steroid sensitivity [17]. In Brazil, carriers of two alleles started dialysis 12 years earlier than patients without risk alleles, a difference that remained at 9 years after adjustment for sex, smoking, body mass index and diabetes [12]. In the North American histopathological series, patients with two alleles who underwent biopsy for arterionephrosclerosis had a mean age of 47 years vs 56 years in patients without risk alleles [3]. In Bahia, progression to end-stage kidney disease increased in a graded manner with the number of risk alleles: 10% with zero alleles, 14% with one and 56% with two (p < 0.001), and sickle cell trait added independent risk, with progression in 39% vs 13% in patients with haemoglobin AA (p = 0.01) [13]. The typical age of onset of end-stage kidney disease in APOL1-associated nephropathy is relatively young, often before 55 years; the risk is greatest at young and middle ages and the age from which it decreases is unknown, it being presumed that older carriers without chronic kidney disease have escaped the modifying factors and are at lower risk, which is why the 2021 Delphi consensus called for age-dependent prediction models incorporating the other risk factors [30].

There is a genetic modifier with a favourable prognostic effect. The presence of the p.(Asn264Lys) variant in cis with a G2 allele reduces the risk to levels comparable to those of non-risk genotypes, with a 93% reduction in the odds ratio for focal segmental glomerulosclerosis and no cases among G2/G2 carriers of the variant; in a patient with chronic kidney disease, a high-risk genotype and this variant, the probability that the disease has another cause is high, an alternative cause having been identified in 76.9% of these cases in a biobank cohort [16,19].

The extrarenal prognosis does not appear to be worsened by the genotype: the individual-participant meta-analysis of 21,305 Black people found no association with incident cardiovascular disease (hazard ratio 1.11; 95% CI 0.96–1.28) or with mortality, regardless of kidney function, age, diabetes and body mass index [34].

In transplantation and donation, prognosis follows the genotype of the kidney and not that of the person. Kidneys from deceased donors with a high-risk genotype fail faster, regardless of donor ethnicity, and survival of grafts from donors of African ancestry without variants is similar to that of donors without that ancestry [1,48]. In living donors, at a median of 18.5 years after donation, Black donors with a high-risk genotype had a relative risk of 2.31 (95% CI 1.16–4.61) of an estimated glomerular filtration rate below 45 ml/min/1.73 m², attenuated to 1.91 (95% CI 0.90–4.03) after adjustment for pre-donation kidney function; across all 653 donors, 7.0% had this degree of reduction in kidney function [47]. For context, the general natural history of donation indicates an immediate loss of about 50% of kidney function with recovery to about 70% of the initial value, proteinuria above 300 mg per day in up to 12% of donors at seven years, and, at ten years, an estimated glomerular filtration rate between 60 and 80 ml/min/1.73 m² in 10%, between 30 and 59 in 12% and below 30 in 0.2%, with an estimated risk of chronic kidney disease after donation of 0.24% in Black men and 0.15% in Black women [48].

11.2. Surveillance table

There is no guideline-established specific surveillance programme for this entity: the consensus states that there are no established treatments, that the genotype is not in itself actionable and that one of the possible uses of the result is to inform risk stratification and the frequency of follow-up, without setting intervals; the availability of chronic kidney disease care and screening is, moreover, one of the conditions for offering the test [1]. The 2021 Delphi consensus also accepted that the genotype may guide monitoring of progression, without setting tests or intervals [30]. The elements that follow combine chronic kidney disease surveillance, which is the framework into which these patients fit, with the surveillance points specific to the entity and with the monitoring requirements of the drugs used, and each row indicates the source of its content.

11.3. What is needed to improve prognosis

The gaps have been systematised by the consensus and directly constrain the quality of the follow-up that is possible. Longitudinal data on progression by genotype are lacking, including in high-risk carriers who do not progress; prevalence data in children, young adults and pregnant women are lacking; it is not known whether sodium-glucose cotransporter 2 inhibitors, glucagon-like peptide 1 receptor agonists and dual angiotensin and endothelin receptor antagonists alter progression in this form of disease; the prognostic significance of a slightly reduced glomerular filtration rate in high-risk living donors without proteinuria or hypertension remains to be determined; and biomarker panels that, coupled with kidney failure risk equations, would allow identification of those who will progress remain to be established. Biomarkers with an independent association with the renal outcome include tumour necrosis factor receptors 1 and 2 and kidney injury molecule 1, whose joint elevation increases the risk [1].

The clinical question that dominates the near horizon is whom to treat, when treatment exists: whether to treat in the presence of significant podocytopathy with high proteinuria, whether to treat moderate progression in younger people, or whether to treat on the basis of validated prediction equations and biomarker panels. This is, verbatim, the formulation of the research agenda, and the transformation of current follow-up, essentially observational, into follow-up with therapeutic consequences depends on it [1,57].

12. Summary and Final Considerations

APOL1-associated kidney disease occupies a place of its own in medical genetics because it is neither a monogenic disease nor a simple polygenic risk factor. Two common coding variants in the last exon of APOL1 — the G1 haplotype, composed of c.1024A>G, p.(Ser342Gly), and c.1152T>G, p.(Ile384Met), and the G2 deletion, c.1164_1169del, p.(Asn388_Tyr389del) — confer, when inherited in two copies, a large increase in the risk of various forms of kidney disease, through a mechanism of toxic gain of function of the protein's channel activity, and not of deficiency. The variants reached high frequencies in sub-Saharan Africa because they protect against the trypanosomes that cause sleeping sickness, and the African diaspora carried them to the Americas and Europe [2,5,27].

The essential point for clinical practice is the pairing of effect magnitude and low penetrance. In African Americans, the recessive odds ratio is 17 for focal segmental glomerulosclerosis and 29 for HIV-associated nephropathy, and in antiretroviral-naive South Africans it reached 89 for HIV-associated nephropathy; at the same time, about 15% of carriers of a high-risk genotype will develop kidney failure and 5% to 8% focal segmental glomerulosclerosis, with an estimated lifetime risk of about 4% in the absence of a known second hit [1,16,17,36]. What determines the outcome is the existence and size of the second hit, high-interferon states and viral infections being the best documented — HIV infection, COVID-19 and, more recently, dengue —, which is why HIV testing is mandatory in any compatible nephropathy and why treatment of the infection is the intervention with the greatest demonstrated impact [1,6,29,37].

The clinical spectrum is broad and cuts across the histopathological classification: focal segmental glomerulosclerosis, collapsing glomerulopathy, HIV- and COVID-19-associated nephropathy, chronic kidney disease attributed to hypertension, accelerated progression of lupus nephritis, sickle cell nephropathy and, in pregnancy, a risk of pre-eclampsia linked to the fetal genotype. The slowly progressive subnephrotic form is the most common and the most frequently misclassified; biopsy may recognise it by the combination of little obsolescent glomerulosclerosis with solidified glomerulosclerosis, thyroidisation-type tubular atrophy and microcystic tubular dilatation, with a specificity of 86% to 92% but a sensitivity below 45% [1,3,13,45].

The diagnosis rests on two pieces of information that do not replace each other. The genotype is definitive as to the presence of the alleles and says nothing about the cause of a particular individual's disease; histology characterises the lesion and does not identify the genotype. Genotyping is no longer acceptable without the protective modifier p.(Asn264Lys), which, in cis with G2, practically abolishes the risk; its presence in a patient with chronic kidney disease and a high-risk genotype requires another cause to be sought, identified in more than three-quarters of cases in a biobank cohort. The phase between G2 and the modifier matters, because the demonstrated protection exists in cis and not in trans [1,16,19,20,32].

The decision to test is, in this entity, a decision with an explicit ethical dimension. At population level there is no evidence to support screening, because there are no established treatments and the genotype is not in itself actionable; at individual level, testing is justified when cumulative conditions are met: consent, membership of an at-risk population, a relevant clinical situation, utility for management, proportionality, availability of qualified counselling and, possibly, relief of anxiety or reproductive information. The correct formulation is to ask about African ancestry and not about race, and status is determined by genotyping and not by appearance or self-identification. The testing policy panel, made up mostly of Black participants, rejected both the absence of information and consent and the imposition of testing and the prohibition of donation on the basis of genotype [1,6,55]. The 2021 Delphi consensus recommended considering the entity in every kidney patient of African ancestry and in any kidney patient with a high-risk relative, assigned to the nephrologist the counselling, ordering and return of the result, and set out the content of pre- and post-test counselling, with 55 statements supported mainly by low- or very-low-quality evidence [30].

The clinical scenario in which the genotype currently changes a decision is living kidney donation. The donor cohort with a median follow-up of 18.5 years showed a relative risk of 2.31 of an estimated glomerular filtration rate below 45 ml/min/1.73 m² in Black donors with a high-risk genotype, and the recommendations range from «may be offered» in the KDIGO living donor guideline to «offer to all potential donors of African heritage under 60 years» from the British transplantation society, via the North American recommendation to offer counselling and testing to all candidates of sub-Saharan or West African ancestry, regardless of family history. On the graft side, the genotype of the donor kidney, and not donor ethnicity, is the determinant of faster failure, with consequences under discussion for allocation indices [1,20,47,48].

There is no authorised targeted therapy. Treatment is that of the proteinuric kidney disease present, with the caveat that most progression data predate the widespread use of modern nephroprotective therapies and that their effect in this specific form has not been established; the doses used are extrapolated from labels for related indications. Four strategies are in development: direct channel blockade, with inaxaplin reducing proteinuria by 47.6% in a single-arm study with thirteen patients assessed and, in an uncontrolled phase 2b study, by 42.7% in a cohort with modest proteinuria but only 17.3%, with an interval crossing zero, in the cohort with type 2 diabetes; reduction of expression by antisense oligonucleotide; inhibition of Janus kinase pathway-dependent induction, whose trial closed with eight participants and no results; and downstream anti-inflammatory approaches. The pivotal trial, with 522 participants and an interim analysis expected in early 2027, has as its final outcome the slope of estimated glomerular filtration rate after two years, and no trial has so far reported a hard clinical outcome. A concern specific to this entity is the effect of blocking the protein on the risk of trypanosomiasis, which should be monitored in trials [1,6,21,22,40,41,54].

It remains to emphasise what has changed in the way this disease is counted. The adoption of the term «APOL1-associated kidney disease» and the creation of dedicated codes in the US international classification of diseases correct a problem that was not one of nomenclature: the absence of a codable diagnosis forced the misclassification of these patients, confused the epidemiology of kidney failure and delayed research. With almost a third of the West African population having a high-risk genotype and more than one hundred million people in sub-Saharan Africa possibly in that situation, the potential magnitude of the problem is large and prevalence data are still lacking in several diaspora populations, including Portugal, where no studies were identified. The entity illustrates, finally, a general principle: a common, low-penetrance risk genotype is not a diagnosis, but it may be the missing information to explain a disease, to anticipate progression and to decide on a donation [1,4,7].

13. References

1. Ojo AO, Adu D, Bramham K, Freedman BI, Gbadegesin RA, Ilori TO, et al. APOL1 kidney disease: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2025;108(5):763-779. doi:10.1016/j.kint.2025.05.017
2. Genovese G, Friedman DJ, Ross MD, Lecordier L, Uzureau P, Freedman BI, et al. Association of trypanolytic ApoL1 variants with kidney disease in African Americans. Science. 2010;329(5993):841-845. doi:10.1126/science.1193032
3. Larsen CP, Beggs ML, Saeed M, Ambruzs JM, Cossey LN, Messias NC, et al. Histopathologic findings associated with APOL1 risk variants in chronic kidney disease. Mod Pathol. 2015;28(1):95-102. doi:10.1038/modpathol.2014.92
4. Freedman BI, Gillespie BS, Egbuna OI. ICD-10 codes for APOL1-mediated kidney disease. Kidney Int. 2026;109(2):238-241. doi:10.1016/j.kint.2025.09.025
5. Online Mendelian Inheritance in Man. *603743 APOLIPOPROTEIN L1; APOL1 [Internet]. Baltimore: Johns Hopkins University; entry updated on 13 February 2025. https://www.omim.org/entry/603743
6. GeneGuidance. Record of the APOL1 gene and mapped conditions (local version of the geneguidance-FULL dump of 26 September 2026).
7. Gbadegesin RA, Ulasi I, Ajayi S, Raji Y, Olanrewaju T, Osafo C, et al. APOL1 bi- and monoallelic variants and chronic kidney disease in West Africans. N Engl J Med. 2025;392(3):228-238. doi:10.1056/NEJMoa2404211
8. Online Mendelian Inheritance in Man. #612551 FOCAL SEGMENTAL GLOMERULOSCLEROSIS 4, SUSCEPTIBILITY TO; FSGS4 [Internet]. Baltimore: Johns Hopkins University; entry edited on 28 September 2022. https://www.omim.org/entry/612551
9. Orphanet. Hereditary steroid-resistant nephrotic syndrome. ORPHA:656 [Internet]. Updated on 2 July 2026. https://www.orpha.net/en/disease/detail/656
10. Lipska-Ziętkiewicz BS. Genetic Steroid-Resistant Nephrotic Syndrome Overview. In: GeneReviews® [Internet]. Seattle: University of Washington; 26 August 2021. NBK573219. https://www.ncbi.nlm.nih.gov/books/NBK573219/
11. International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). N07.B Hereditary nephropathy, not elsewhere classified with APOL1-mediated kidney disease [AMKD]; Z84.11 Family history of APOL1-mediated kidney disease [AMKD] [Internet]. https://icd-10codes.com/code/N07.B
12. Riella C, Siemens TA, Wang M, Campos RP, Moraes TP, Riella LV, et al. APOL1-associated kidney disease in Brazil. Kidney Int Rep. 2019;4(7):923-929. doi:10.1016/j.ekir.2019.03.006
13. Alladagbin DJ, Lima VBF, Tavares MB, Amorim Boa Sorte TRS, Khouri NA, da Luz Neto ER, et al. APOL1 risk genotypes and sickle cell trait are associated with kidney disease progression in an Afro-Brazilian population. Kidney360. 2026;7(2):e0000001378. doi:10.34067/KID.0000001378
14. de Souza CMB, de Oliveira Fam BS, Giudicelli GC, de Menezes Neves PDM, Veronese FV, Vianna FSL. An integrative review of APOL1 kidney disease with a focus on the Brazilian population. Clinics (Sao Paulo). 2026;81:101109. doi:10.1016/j.clinsp.2026.101109
15. Servais A, Gribouval O, Gaillard F, Antignac C. APOL1 risk genotype in Europe: data in patients with focal segmental glomerulosclerosis and after renal transplantation. Nephrol Ther. 2019;15 Suppl 1:S89-S94. doi:10.1016/j.nephro.2019.02.005
16. Martinelli E, Ke J, Khan A, Wongboonsin J, Vanderwall DR, Lim TY, et al. Precision diagnosis in APOL1 kidney disease with the p.N264K M1 protective variant. JAMA Netw Open. 2026;9(3):e261452. doi:10.1001/jamanetworkopen.2026.1452
17. Kopp JB, Nelson GW, Sampath K, Johnson RC, Genovese G, An P, et al. APOL1 genetic variants in focal segmental glomerulosclerosis and HIV-associated nephropathy. J Am Soc Nephrol. 2011;22(11):2129-2137. doi:10.1681/ASN.2011040388
18. Ekulu PM, Nkoy AB, Betukumesu DK, Aloni MN, Makulo JRR, Sumaili EK, et al. APOL1 risk genotypes are associated with early kidney damage in children in sub-Saharan Africa. Kidney Int Rep. 2019;4(7):930-938. doi:10.1016/j.ekir.2019.04.002
19. Gbadegesin R, Martinelli E, Gupta Y, Friedman DJ, Sampson MG, Sanna-Cherchi S. APOL1 genotyping is incomplete without testing for the protective M1 modifier p.N264K variant. Glomerular Dis. 2024;4(1):43-48. doi:10.1159/000537948
20. Ames EG, Anand PM, Bekheirnia MR, Doshi MD, El Ters M, Freese ME, et al. Evaluation for genetic disease in kidney transplant candidates: a practice resource. Am J Transplant. 2025;25(2):237-249. doi:10.1016/j.ajt.2024.10.019
21. Egbuna O, Zimmerman B, Manos G, Fortier A, Chirieac MC, Dakin LA, et al. Inaxaplin for proteinuric kidney disease in persons with two APOL1 variants. N Engl J Med. 2023;388(11):969-979. doi:10.1056/NEJMoa2202396
22. Vertex Pharmaceuticals. Vertex announces positive results from the phase 2b AMPLIFIED study of inaxaplin in additional populations of people with APOL1-mediated kidney disease (AMKD) and completion of enrollment in the phase 2/3 AMPLITUDE trial [press release]. Boston: Vertex Pharmaceuticals; 22 September 2026. https://investors.vrtx.com/news-releases/news-release-details/vertex-announces-positive-results-phase-2b-amplified-study
23. European Rare Kidney Disease Reference Network (ERKNet) [Internet]. European Reference Network funded by the EU4Health programme 2021-2027. https://www.erknet.org
24. NephCure. Institutional website of the patient organisation dedicated to rare proteinuric kidney diseases [Internet]. https://nephcure.org
25. Freedman BI, Moxey-Mims MM, Alexander AA, Astor BC, Birdwell KA, Bowden DW, et al. APOL1 Long-term Kidney Transplantation Outcomes Network (APOLLO): design and rationale. Kidney Int Rep. 2020;5(3):278-288. doi:10.1016/j.ekir.2019.11.022
26. Tzur S, Rosset S, Shemer R, Yudkovsky G, Selig S, Tarekegn A, et al. Missense mutations in the APOL1 gene are highly associated with end stage kidney disease risk previously attributed to the MYH9 gene. Hum Genet. 2010;128(3):345-350. doi:10.1007/s00439-010-0861-0
27. Friedman DJ, Pollak MR. APOL1 and kidney disease: from genetics to biology. Annu Rev Physiol. 2020;82:323-342. doi:10.1146/annurev-physiol-021119-034345
28. Parsa A, Kao WH, Xie D, Astor BC, Li M, Hsu CY, et al. APOL1 risk variants, race, and progression of chronic kidney disease. N Engl J Med. 2013;369(23):2183-2196. doi:10.1056/NEJMoa1310345
29. Wu H, Larsen CP, Hernandez-Arroyo CF, Mohamed MMB, Caza T, Sharshir M, et al. AKI and collapsing glomerulopathy associated with COVID-19 and APOL1 high-risk genotype. J Am Soc Nephrol. 2020;31(8):1688-1695. doi:10.1681/ASN.2020050558
30. Freedman BI, Burke W, Divers J, Eberhard L, Gadegbeku CA, Gbadegesin R, et al. Diagnosis, education, and care of patients with APOL1-associated nephropathy: a Delphi consensus and systematic review. J Am Soc Nephrol. 2021;32(7):1765-1778. doi:10.1681/ASN.2020101399
31. National Center for Biotechnology Information. ClinVar: NM_003661.4(APOL1):c.1024A>G p.(Ser342Gly) VCV000277678; c.1152T>G p.(Ile384Met) VCV000127198; G1 haplotype VCV000006080; c.1164_1169del p.(Asn388_Tyr389del) VCV000006081; compound genotype G1/G2 VCV001120175; c.792C>A p.(Asn264Lys) VCV001170737 [Internet]. Bethesda: National Library of Medicine. https://www.ncbi.nlm.nih.gov/clinvar/
32. Greenberg M, Tabachnikov O, Ben-Ruby D, Vivante A, Skorecki K, Magen D. Case report of a child with kidney disease: consideration of the risk of a single APOL1 G2 allele with a protective N264K variant on the G0 parental chromosome. BMC Nephrol. 2026;27:384. doi:10.1186/s12882-026-05006-9
33. Narjoz C, Tran VHLJ, Rabant M, Karras A, Palich R, Thervet E, et al. Diagnostic yield of APOL1 p.N264K variant screening in daily practice. Kidney Int Rep. 2024;9(6):1915-1918. doi:10.1016/j.ekir.2024.04.008
34. Grams ME, Surapaneni A, Ballew SH, Appel LJ, Boerwinkle E, Boulware LE, et al. APOL1 kidney risk variants and cardiovascular disease: an individual participant data meta-analysis. J Am Soc Nephrol. 2019;30(10):2027-2036. doi:10.1681/ASN.2019030240
35. Palmer ND, Freedman BI. APOL1-mediated and Mendelian forms of "heretofore" idiopathic collapsing glomerulopathy: lessons from Brazil. Kidney Int. 2024;105(4):684-686. doi:10.1016/j.kint.2023.12.014
36. Kasembeli AN, Duarte R, Ramsay M, Mosiane P, Dickens C, Dix-Peek T, et al. APOL1 risk variants are strongly associated with HIV-associated nephropathy in black South Africans. J Am Soc Nephrol. 2015;26(11):2882-2890. doi:10.1681/ASN.2014050469
37. Neves PDMM, Silva LCS, Pereira MEVC, Oliveira FAM, Araújo SA, Wanderley DC, et al. Dengue infection as a trigger for collapsing glomerulopathy in patients with APOL1 high-risk genotype. Kidney Int Rep. 2025;10(11):3741-3745. doi:10.1016/j.ekir.2025.08.005
38. Chang JH, Husain SA, Santoriello D, Stokes MB, Miles CD, Foster KW, et al. Donor's APOL1 risk genotype and "second hits" associated with de novo collapsing glomerulopathy in deceased donor kidney transplant recipients: a report of 5 cases. Am J Kidney Dis. 2019;73(1):134-139. doi:10.1053/j.ajkd.2018.05.008
39. ClinicalTrials.gov. Phase 2a study of VX-147 in adults with APOL1-mediated focal segmental glomerulosclerosis. NCT04340362 [Internet]. Updated on 24 September 2025. https://clinicaltrials.gov/study/NCT04340362
40. ClinicalTrials.gov. Phase 2/3 adaptive study of VX-147 in adult and pediatric participants with APOL1-mediated proteinuric kidney disease (AMPLITUDE). NCT05312879 [Internet]. Updated on 26 October 2026. https://clinicaltrials.gov/study/NCT05312879
41. ClinicalTrials.gov. Janus kinase-STAT inhibition to reduce APOL1-associated kidney disease (JUSTICE). NCT05237388 [Internet]. Updated on 24 April 2026. https://clinicaltrials.gov/study/NCT05237388
42. Varner JD, Ilori TO, Gbadegesin RA. Update on APOL1 and chronic kidney diseases in children. Pediatr Nephrol. 2026;41(11):3813-3826. doi:10.1007/s00467-026-07384-6
43. Adeyemo A, Esezobor C, Solarin A, Abeyagunawardena A, Kari JA, El Desoky S, et al. HLA-DQA1 and APOL1 as risk loci for childhood-onset steroid-sensitive and steroid-resistant nephrotic syndrome. Am J Kidney Dis. 2018;71(3):399-406. doi:10.1053/j.ajkd.2017.10.013
44. Vendrig LM, Ke J, Tanck MWT, Lim TY, Martinelli E, Bodria M, et al. APOL1 kidney risk variants and outcomes in children with congenital anomalies of the kidney and urinary tract. Pediatr Nephrol. 2026;41(4):1123-1132. doi:10.1007/s00467-025-07127-z
45. Reidy KJ, Hjorten RC, Simpson CL, Rosenberg AZ, Rosenblum SD, Kovesdy CP, et al. Fetal-not maternal-APOL1 genotype associated with risk for preeclampsia in those with African ancestry. Am J Hum Genet. 2018;103(3):367-376. doi:10.1016/j.ajhg.2018.08.002
46. Afsar B, Afsar RE, Lentine KL. The relationship between APOL1 risk variants and preeclampsia and fetal outcomes. Pregnancy Hypertens. 2026;43:101458. doi:10.1016/j.preghy.2026.101458
47. Hsu CY, Gao Y, Freedman BI, Lunn MR, Muiru AN, Schnitzler MA, et al. Apolipoprotein L1 gene genotype and kidney outcomes after living kidney donation. JAMA Intern Med. 2026;186(8):943-950. doi:10.1001/jamainternmed.2026.0996
48. Tavares MG, Requião-Moura L, Queiroga L, Silva Junior HT, Pestana JOM. The APOL1 gene and kidney transplantation: a review article. J Bras Nefrol. 2026;48(2):e20250093. doi:10.1590/2175-8239-JBN-2025-0093en
49. Eadon MT, Cavanaugh KL, She L, Steen-Burrell KA, Mohottige D, Nadkarni GN, et al. Genetic testing for APOL1 in adults with hypertension: the GUARDD-US randomized clinical trial. JAMA Netw Open. 2026;9(3):e260528. doi:10.1001/jamanetworkopen.2026.0528
50. Nadkarni GN, Fei K, Ramos MA, Hauser D, Bagiella E, Ellis SB, et al. Effects of testing and disclosing ancestry-specific genetic risk for kidney failure on patients and health care professionals: a randomized clinical trial. JAMA Netw Open. 2022;5(3):e221048. doi:10.1001/jamanetworkopen.2022.1048
51. van Schijndel E, Roelofs J, Hilhorst M, Bemelman F. Collapsing FSGS in a kidney transplant with a single APOL1 risk variant: a case report. BMC Nephrol. 2026;27:142. doi:10.1186/s12882-026-04789-1
52. U.S. Food and Drug Administration. Losartan potassium tablets, for oral use: highlights of prescribing information. ANDA 078243; label version with effective date 27 August 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
53. U.S. Food and Drug Administration. FARXIGA (dapagliflozin) tablets, for oral use: highlights of prescribing information. NDA 202293. AstraZeneca; label version with effective date 10 July 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
54. U.S. Food and Drug Administration. FILSPARI (sparsentan) tablets, for oral use: highlights of prescribing information. NDA 216403. Travere Therapeutics; label version with effective date 9 April 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
55. McIntosh T, Walsh H, Baldwin K, Iltis A, Mohan S, Sawinski D, et al. Evaluating ApoL1 genetic testing policy options for transplant centers: a Delphi consensus panel project with stakeholders. Clin J Am Soc Nephrol. 2024;19(4):494-502. doi:10.2215/CJN.0000000000000397
56. Iltis AS, Walsh HA, Baldwin K, McIntosh T, Mohan S, Sawinski D, et al. Insights from Black living kidney donors: an interview study on APOL1 genetic testing experiences. Clin Transplant. 2026;40(2):e70482. doi:10.1111/ctr.70482
57. Hopper T, Wang B, Olabisi OA. APOL1-mediated kidney disease and the emerging era of precision therapy. Annu Rev Med. 2026;77. doi:10.1146/annurev-med-042325-051210
58. ClinicalTrials.gov. A phase 2 study of MZE829 in adults with APOL1 kidney disease. NCT06830629 [Internet]. Updated on 27 July 2026. https://clinicaltrials.gov/study/NCT06830629
59. ClinicalTrials.gov. A study to assess safety, tolerability, pharmacokinetics and pharmacodynamics of AZD2373 in healthy male participants of sub-Saharan West African ancestry. NCT05351047 [Internet]. Updated on 7 August 2023. https://clinicaltrials.gov/study/NCT05351047
60. Egbuna O, Audard V, Manos G, Tian S, Hagos F, Chertow GM. Safety and tolerability of the APOL1 inhibitor, inaxaplin, following single- and multiple-ascending doses in healthy adults. Glomerular Dis. 2024;4(1):61-71. doi:10.1159/000538255
61. ClinicalTrials.gov. Dose-ranging safety, tolerability, and efficacy study of AZD2373 in participants with APOL1-mediated kidney disease (APPRECIATE). NCT06824987 [Internet]. Updated on 13 October 2026. https://clinicaltrials.gov/study/NCT06824987
62. ClinicalTrials.gov. Inaxaplin in participants with proteinuric APOL1-mediated kidney disease with or without comorbidities (AMPLIFIED). NCT06794996 [Internet]. Updated on 20 August 2026. https://clinicaltrials.gov/study/NCT06794996
63. Elsurer Afsar R, Clair B, Memon AA, Edwards JC, Miyata KN, Afsar B, et al. Beyond the test: patient-reported concerns and worries regarding kidney conditions and APOL1 genetic testing among Black patients. Front Nephrol. 2026;6:1815147. doi:10.3389/fneph.2026.1815147